Massively Densified Networks Why We Need Them and How We Can Build Them Monica Paolini, Senza Fili

Massively densified networks Why we need them and how we can build them Monica Paolini, Senza Fili

We thank these companies for sponsoring this report: In collaboration with

I. Report: Moving to dense, pervasive networks 3 II. Vendor profiles and interviews 29 1. Densification is more than small cells or DAS or C-RAN. From Anritsu 30 atomic to pervasive networks 4 Ascom Network Testing 37 2. The emerging RAN taxonomy. Antenna coverage and baseband CommScope 44 separation 8 InterDigital 52 3. The big question: Indoor / Outdoor 10 Kathrein 59 4. Drivers: Coverage / Capacity 11 Rohde & Schwarz 66 5. Performance: Capacity / Latency 12 Samsung Networks 74 6. Architecture: Small cells / DAS 13 SOLiD 81 8. Network: Distributed / Centralized 15 SpiderCloud Wireless 88 9. Technology: Cellular / Wi-Fi 17 III. Operators’ interviews 95 10. Unlicensed spectrum: LTE / Wi-Fi 18 11. Spectrum: Sub-6 GHz / Millimeter wave 20 BT 96 12. Interference: Co-channel / Separate channel 21 DOCOMO Innovations 104 13. Backhaul: Fiber / Wireless 22 Carolina Panthers 110 14. Fronthaul: CPRI / Xhaul 23 Enterprise, anonymous 116 15. Access point density: High / Low 25 Glossary 120 16. Business model: Single operator / Shared deployment 26 References 122 17. Concluding thoughts. The role of operators in a densified, pervasive network 28

We often equate densification with the small-cell deployments, usually in conjunction with Wi-Fi and DAS, that increase wireless network capacity to accommodate traffic growth. While small-cell deployments are certainly a central element in the densification process, densification itself is becoming the catalyst for a much deeper evolution in wireless networks, which is not limited to small-cell deployments, and which is enabled by demand drivers and new technologies but goes beyond the contribution that each of them separately brings.

Today’s networks have an atomic, discrete architecture in which cells are the edge access elements and are all connected to a distinct, common core. In the initial stages of densification, operators increase the number and density of these elements with a surgically precise addition of small cells, DAS or carrier Wi-Fi elements, but the architecture remains fundamentally the same. As this process intensifies, these atomic networks start to change into what we call pervasive networks in this report. Others (notably among them, I Chih-Lin at China Mobile) have called them user-centric networks, as opposed to the traditional network- centric network.

Pervasive networks are distributed. Edge elements of the network, such as small cells or DAS, get closer to users and devices. And devices themselves can become part of the access network itself, with device-to-device connectivity.

With C-RAN and, more generally, virtualization, the cell as the fundamental self- contained element in the RAN ceases to exist. It is replaced by a multi-layer, multiband set of antennas connected to a remote baseband. Devices within this model can connect to more than one antenna and, in a mobility scenario, switch from one antenna to the next without having to do a handoff, because the cell ID remains the same. This is the evolution model put forward by China Mobile’s no-more-cells approach and DOCOMO’s phantom cells.

REPORT Massively densified networks © 2016 Senza Fili www.senzafiliconsulting.com |4| Demand is the main driver to densification. Traffic growth on wireless networks continues unabated as subscribers conduct more activities and a wider variety of them, over more mobile devices, and as IoT spreads. In turn, the increased coverage and capacity of densified wireless networks facilitate subscriber and IoT device access, and this drives a further increase in demand. Increasing network capacity to meet demand has become financially unsustainable in today’s atomic networks. Pervasive networks allow a more efficient use of network resources that will enable operators to provide the capacity and performance needed cost-efficiently.

Outdoor small cells were the first solution to address the need for densification as a complement to Wi-Fi offload. Small cells can be deployed in both 3G and 4G networks, and can be combined with Wi-Fi, sharing spectrum with the macro layer. As operators started to test small cells and plan for deployments, though, they realized that moving to large-scale deployments of small cells required substantial operational and financial effort.

Over the last few years, the entire wireless ecosystem has been working to find business models and technological solutions that meet the operators’ performance and cost requirements. The rest of the report will discuss the advancements in this area and how they relate to pervasive networks in detail. For now, we can call out 5G and virtualization as the crucial technology enablers in the transition to pervasive networks. The two technologies combine the performance improvement, the cost effectiveness, and the flexibility that operators need to meet the growth in demand in their networks.

The transition from atomic to pervasive networks has a major impact on wireless networks – from technology, performance, usage model, and financial perspectives, as described in the table below. The evolution is not confined to wireless technology or network architecture. It affects the entire ecosystem, including subscribers, enterprise, and third-party players, as well as business, ownership, and operational models.

Densification is necessary for wireless networks to meet demand, but many changes, are necessary to achieve the capacity and performance goals, and they will eventually transform today’s atomic networks into pervasive networks.

Atomic networks Pervasive networks

Network model Network-centric: subscriber adapts to the network (e.g., goes to the User-centric: network adapts to subscriber demand (e.g., ultra-dense window to make a phone call). wireless infrastructure in stadiums). RAN Discrete elements: cells (antenna and baseband). No-more-cells, phantom-cells approach, with antennas as access points in a multi-layer topology, connected to a remote baseband. UE-RAN connection One-to-one connection from the UE to the cell. UEs can connect to multiple antennas, use multiple bands. Handoffs required for the UE to move association from one cell to Flexible modes of connectivity coexist: dual connectivity, device-to- the next. device connectivity, Wi-Fi offload. Subscriber can establish multiple concurrent connections: multiple devices (including non-SIM and IoT devices) on the same plan. Distinction between RAN elements and devices is less sharp because devices connect to each other and act as access points to the RAN. User and control User and data planes allocated to each access channel (e.g., sector). Control plane can manage traffic for multiple access channels, so planes some access channels do not require a separate control plane (e.g., LTE in unlicensed bands, LWA, mmW). Short-range mobility can be managed without handoffs. Densification targets Coverage in the wide area, capacity in high-traffic areas, with most of Vertical capacity increase and coverage extension driven by location- the RAN infrastructure in outdoor locations and large venues (e.g., specific traffic or service requirements (e.g., service tied to a venue; stadiums). IoT service). Layers Single macro-layer, possibly with limited small-cell hotspot Multi-layer networks, with extensive indoor and outdoor coverage deployments, and with Wi-Fi offload. with small cells, DAS or femto cells. Spectrum Cellular frequencies below 3 GHz. Wider range of higher-frequency bands (3.5 GHz, 5 GHz, mmW), with the inclusion of unlicensed spectrum. Core/RAN Separate location and equipment, with RAN equipment located at Boundaries less strict, with RAN becoming virtualized and the edge and core equipment in centralized locations. centralized, and some wireless core functionality moving to the edge (e.g., MEC, CORD). Location of function (distributed versus centralized architectures) is a strategic decision.

Testing, monitoring, Testing and monitoring based on network KPIs and historical data. QoE metrics based on performance of UEs are tied to network KPIs optimization Limited optimization functionality. to test, monitor and optimize networks in real time.

Performance yardstick Capacity per RAN element. Capacity density (e.g., per sq km) and latency.

Traffic management Maximize throughput. Real-time traffic management, at the application or service level. Capacity determines service availability. Network slicing used to extract more value from network resources. RAN equipment Telecom assets (e.g., macro-cellular towers, building rooftops), RAN equipment gets closer to subscribers and devices – closer to the location mostly in outdoor locations. ground and indoors.

Network ownership Operator owns network, often leasing space on cell tower or other Venue owners increasingly pay for infrastructure, even though they assets. Limited network sharing. do not (and choose not to) operate the network. Multi-operator, neutral-host model, in which some network elements (e.g., backhaul) are shared among operators. Control Operators control end-to-end network. Operators retain control of the RAN, but other players – venue owners, residential users, neutral hosts and system integrator – get more visibility into the networks and have a stronger role in determining how the network resources they paid for are being used.

Densification dictates an evolution of the RAN that involves all its elements. For operators, densification for capacity purposes starts with the macro networks. When operators need more capacity in an area, they typically first densify the macro network where possible, by adding cell sites, splitting sectors or adding new channels or bands. At some point this becomes financially too expensive or difficult (e.g., in environments where antenna density is already high, or where it is difficult to find new cell sites), and operators move densification to sub-layers – micro cells, small cells or femto cells. DAS deployments typically run in parallel, to address high-density venues such as stadiums or enterprises.

Distinguishing among different RAN elements, from macro cells to femto cells, including DAS, has become increasingly difficult. Many solutions are available to address specific environments, and do not neatly fit into any of the traditional RAN element groups. Rather, there is a continuum of solutions that are needed jointly, to address the varying requirements of different environments. This is a welcome evolution that testifies to the increased awareness of the multitude of environments where we need further densification, and the specific challenges that each presents.

At the same time, the trend toward C-RAN and, more generally, toward the virtualization of the RAN creates a second dimension by which to define RAN elements. The first dimension is the antenna coverage area, which decreases in the move from macro to femto cells, but without well-defined borders among the different element types. The new, second dimension is the location of the baseband. In a distributed, traditional RAN, the baseband is at the cell site at the edge. In a centralized network, the baseband is located remotely. There are different types of baseband separation, depending on the type of fronthaul used – and the type of functional split that defines different types of xhaul.

REPORT Massively densified networks © 2016 Senza Fili www.senzafiliconsulting.com |8| Like macro cells, small cells can be distributed or centralized. Small cells and C-RAN are often presented as alternatives to each other, and small cells are seen as competing with DAS. However, within the frame of the densification process, small cells and DAS converge to a set of solutions with varying degrees of centralization – with DAS being always centralized, and small cells being either centralized or distributed. As a result, we end up with a continuum of solutions on both axes. This helps operators find a solution that is well-tailored to their needs and helps vendors create differentiation in the marketplace – but it also increases the complexity of the selection process.

Mobile phones were initially developed to provide the wide-area connectivity that wireline phones could not provide. The first mobile phones used satellites for access and were car phones, and there was widespread doubt that mobile phones could be of any use in areas like Manhattan’s urban canyons or for more than short calls.

Today the situation is reversed. Most traffic – 80% or more in most markets – comes from indoor locations, and over 90% is data. Yet most of the RAN infrastructure, if we exclude Wi-Fi, is located outdoors; as a result, indoor coverage and capacity are more limited than outdoor. As wireless has become the default communication interface, the ability to provide the same level of service indoors and outdoors becomes a high-priority requirement for mobile operators – and one that, in most locations, cannot be met cost-effectively with only outdoor RAN infrastructure taking an outside-in approach.

Wi-Fi has been a boon for mobile operators: it transports the bulk of traffic from mobile devices, and most of that traffic is from indoor locations – public venues, enterprise locations, homes. While Wi-Fi continues to complement cellular access, mobile operators want to improve indoor cellular coverage, both to meet the demand from indoor subscribers and to relieve pressure on the macro network. Indoor traffic typically is more expensive to carry than outdoor traffic, because it uses less efficient modulation schemes and hence uses more network resources.

As a result, in recent years, operators have expanded their indoor coverage efforts. With the exception of some Asian countries, particularly Japan and Korea, mobile operators have been cautious about in-building networks, with the exception of large venues such as stadiums. Indoor coverage presents its own challenges, which are markedly different from those in outdoor deployments. The technology, the solutions and the business models are different, and they are evolving along with the relationship among venue owners, operators, and third-parties.

The narrative of wireless network deployments is in many ways all about densification. Mobile network performance and capacity have improved tremendously, thanks to technological innovation and greater spectrum availability, and densification has been a key part of that improvement. Initially densification was used to address coverage holes, or establish more consistent coverage. With the growth in data usage, capacity requirements have become a major driver to densification – and densification has become a top priority for operators.

Increasingly, however, coverage and capacity have intertwined. Just being able to receive a signal at a given location is no longer sufficient for coverage there. Depending on the market, location and operator, the capacity required for adequate coverage varies, and it is becoming meaningless to define coverage without reference to minimum capacity requirements. As a result, a location that was deemed to have coverage in the past may no longer have basic coverage, and it becomes a new densification target in order to achieve sufficient capacity.

At the same time, boosting capacity in a hotspot may improve coverage in the surrounding area. An example is indoor infrastructure that addresses the demand created by indoor user offloads from the macro network serving the location; boosting that hotspot’s capacity also increases the capacity and coverage area of the macro network. In this case, the small cells or DAS in the indoor deployment not only increase the capacity density within their footprint, they also improve performance and coverage in the wider area.

The co-dependency of RAN elements within the same footprint in determining both capacity and coverage demonstrate the need – and benefits – of looking at densification efforts within all the layers of the RAN environment rather than on the RAN elements – e.g., small cells or DAS – that are directly involved.

5. Performance: Capacity / Latency

Capacity and coverage are the drivers to densification: they make it possible for operators to support the services that subscribers pay for. But capacity and coverage are no longer sufficient to make these subscribers happy. QoE, the quintessential – if somewhat elusive and difficult to quantify – measure of subscriber happiness, is increasingly determined not just by service availability (enabled by coverage and capacity), but by latency, as well as other transmission metrics such as jitter and packet loss.

Latency’s rise to prominence is due to the increased use of real-time data applications: streaming video and audio (such as YouTube, Spotify), voice (including VoLTE, OTT voice), entertainment and gaming (e.g., Pokémon). Some IoT applications – e.g., connected cars, safety, monitoring, medical, financial – have tight latency requirements, too.

With real-time applications, poor coverage, congestion and high latency affect QoE in comparable ways: they create a poor subscriber experience, with subscribers giving up on the application they want to use, or with application becoming unavailable. Common effects of high latency include delays on voice calls and games, and, with video, frozen streams, dropped frames, pixelization or long startup times.

As operators plan to increase coverage and capacity via densification, the ability to use low- and, as we move to 5G, ultra-low latency becomes a determinant when choosing the end-to-end network topology. The RAN plays a crucial role in latency, but so do backhaul/fronthaul, transport, core functionality, application and content processing and availability; they all need to be factored into network deployment and optimization for densified networks.

6. Architecture: Small cells / DAS

Small cells and DAS are frequently considered to be competing solutions taking opposed approaches to densification. Historically they have developed to address different requirements: small cells mostly for capacity, DAS mostly for coverage. They generally serve different environments, as well: small cells in outdoor and small-venue / residential locations, DAS for indoor environments and large outdoor venues (although DAS can also be deployed in other outdoor locations).

But DAS and small cells are quickly converging in a varied set of solutions that address the varied needs of venue owners and operators, and that combine features of both small cells and DAS. Both small cell and DAS vendors have increasingly expanded their portfolio to include both solutions or hybrid solutions (e.g., CommScope’s OneCell, Ericssons’s Dot and Huawei’s LampSite).

RAN virtualization pushes small cells even closer to DAS. Furthermore, DAS is a precursor of C-RAN. This is especially true of active DAS topologies that allow a higher level of control over the management of network resources.

An additional push for the convergence comes from the need to address medium- size venues – sometimes referred as the middleprise. Most in-building deployments target large venues because of those locations’ prominence and the dense demand there. Wireless performance during the Super Bowl, for instance, is tracked at an unparalleled depth. Locations like stadiums attract much attention and investment from venue owners and operators.

Smaller venues are a much larger market (e.g., in terms of footage), but are much more challenging to cover profitably, because that market is fragmented and, with some exceptions, smaller venues do not have high capacity requirements or high

REPORT Massively densified networks © 2016 Senza Fili www.senzafiliconsulting.com |13| revenue opportunities. Hybrid solutions and variations on the established DAS and small cell solutions – especially when coupled with some degree of virtualization – are necessary to address mid-size venues. At the same time, the ability to cover mid- size venues is crucial to the transition to massively densified networks, because of the amount of traffic generated by subscribers at these locations.

One of the main decisions operators have to make as they densify their networks is how distributed or centralized the RAN and core functions should be to maximize spectrum and resource utilization, optimize performance, and keep costs down.

The dominant approach to RAN densification in today’s atomic networks – with the exception of DAS – starts with a phase in which densification is distributed: small and femto cells that include all eNB functionality – radio and baseband – are deployed at the edge of the network, often in places where the macro network has insufficient coverage or none at all. This approach makes small and femto cells fast and easy to deploy, because no central location is needed to host baseband functionality. Backhaul requirements can be easily managed because of the lower capacity and latency requirements of a centralized environment.

Centralized architectures, such as DAS, C-RAN and vRAN, present multiple advantages over distributed topologies. Cost benefits that accrue from concentrating baseband processing in a remote location apply to all RAN elements, from macro to small cells. Femto cells may benefit from some level of centralization, but typically the approach to virtualization is different because femto cells do not have cost- effective access to fronthaul. Cost is a key consideration in driving RAN centralization and virtualization of the macro infrastructure in the short term, but the performance and flexibility advantages are going to have a stronger impact in the mid to long term, especially in HetNet multi-layer environments. Cost savings are less for small- cell than for macro-cell deployments, because centralized topologies require fronthaul instead of backhaul to meet latency and capacity requirements, and fronthaul is typically expensive and accounts for a larger percentage of the capex and opex in a small-cell network than in a macro-cell network.

First among the benefits of a centralized architecture is the ability to manage transmission across multiple layers to minimize the effect of interference in co-

REPORT Massively densified networks © 2016 Senza Fili www.senzafiliconsulting.com |15| channel environments. Especially in outdoor environments where macro and small cells overlap in coverage and use the same spectrum channel, interference severely cuts into the capacity benefits of small cells; interference reduces the capacity not only of the small cells, but of the macro cells, which are more expensive and valuable on a per-bit basis.

Indoor small cells and DAS deployments suffer less from interference. The reduced coverage from the macro network, which is what drives in-building deployments, means interference is also more limited. New building codes that shield buildings from macro transmissions further reduce indoor coverage and interference, and indirectly foster a stronger commitment from mobile operators and the enterprise toward indoor deployments. In this environment, a centralized architecture is often beneficial, because it helps manage intra-layer interference, and it makes equipment installation and operation easier.

Centralized deployments also enable – but do not require – new ways to manage traffic in dense environments. Cell IDs can encompass multiple antennas and multiple RATs can be tightly integrated, such as in China Mobile’s no-more-cells model and in DOCOMO’s phantom cells. A centralized, virtualized RAN is better suited to load balancing traffic across antennas and wireless interfaces, and to managing network-sliced traffic, because all the processing is done in a single location where there is full visibility across the real-time load and availability of the locally available network resources.

Finally, a centralized architecture increases the efficiency of instruments like MEC that shift the core functionality to the edge – for instance, to lower latency (e.g., for video streaming), or to support venue-specific applications (e.g., local breakout for enterprise or IoT applications). In this case, deploying MEC in a C-RAN is less expensive and more efficient than in a distributed RAN, because less equipment is needed and better coordination can be achieved. For instance, content caching is more efficient in a centralized environment, where it is available to multiple access locations, than in a distributed RAN in which the caching is done at each access location.

By far the most successful densification strategy to date has been Wi-Fi offload. It is Wi-Fi that has replaced wireline with wireless as the default access technology, not cellular. It was Wi-Fi that showed us what mobile broadband could do. When the iPhone came out, 3G networks did not have sufficient capacity to reliably support bandwidth-intensive services like video streaming, but Wi-Fi did, and it provided many subscribers the motivation to buy the new device.

According to Cisco VNI, Wi-Fi carries 43% of the IP traffic today, and VNI forecasts this percentage to grow to 50% by 2020. By comparison, cellular traffic accounts for 5% today, and that is forecast to be 16% in 2020. This means the dominant access technology for mobile devices is mostly outside operators’ control. Despite the growth in carrier Wi-Fi and, more generally, the increased push for Wi-Fi offload – for data, but also for voice with Wi-Fi Calling – it is the active choice of subscribers that has made Wi-Fi access prevalent, enabled by the wide availability of Wi-Fi infrastructure in residential and enterprise environments.

Wi-Fi is set to continue to play a crucial role as wireless networks densify, but at the same time the 2.4 GHz and 5 GHz spectrums it uses are getting congested – and the congestion will increase with the introduction of LTE access in the 5 GHz band (see below). Wi-Fi will expand to the 60 GHz band, but the dominant use cases there are for services in which the devices are in close proximity to the access point and fundamentally stationary.

As a result, Wi-Fi is likely to become the technology that, instead of providing offload, will need offload itself. An expansion in the allocation of unlicensed spectrum that Wi-Fi may use will bring relief, but we will also need greater availability, higher spectral efficiency and more intensive utilization of cellular bands. Densification can no longer be primarily entrusted to Wi-Fi; it will instead require the integration of multiple access technologies to provide the seamless connectivity that users expect.

Wi-Fi’s success in winning the hearts of wireless users and in using spectrum with unprecedented efficiency, mostly because of dense deployments, has piqued the interest of vendors and operators that have seen an opportunity to use unlicensed spectrum for LTE.

The advantages to mobile operators are clear: operators gain opportunistic – i.e., not guaranteed, contingent on availability – access to unlicensed spectrum, using the same technology and network infrastructure they use for LTE, maintaining control over traffic and integrating it with their cellular RAN and core. In addition, LTE’s spectral efficiency is higher than Wi-Fi’s because of more efficient modulation. With the exception of MulteFire (see below), LTE in unlicensed bands is deployed alongside licensed LTE, and, as a result, the marginal cost of adding LTE unlicensed is low in greenfield small-cell deployments.

There are, however, disadvantages to using LTE instead of Wi-Fi in unlicensed bands as well. Wi-Fi is already installed in virtually all mobile-broadband devices. The infrastructure is widely available, and in most cases free to access – both to operators and to subscribers. In contrast, LTE in unlicensed spectrum requires new devices and new infrastructure that has to be deployed – and paid for, in most cases – by the operator. Besides, in order to use LTE in unlicensed bands without unduly affecting Wi-Fi performance, LTE has to use LBT mechanisms that reduce its performance advantages over Wi-Fi. Finally, deployments of LTE unlicensed require the consent of venue owners where they have control over the location of the LTE unlicensed antennas. Those owners might not grant permission to install, because LTE unlicensed – even when it uses LBT – competes with the local Wi-Fi for network resources, which venue owners consider theirs and want to continue to control.

It is still unclear how widely LAA, the version of LTE unlicensed that is designed to guarantee nice coexistence with Wi-Fi and which is deployable worldwide,

REPORT Massively densified networks © 2016 Senza Fili www.senzafiliconsulting.com |18| worldwiwill be adopted in the face of the competition from Wi-Fi and the LWA allows mobile operators to use unlicensed spectrum but operates in tradeoffs it requires. the opposite direction of LAA. It uses the Wi-Fi air interface, but it fully integrates the Wi-Fi traffic within the LTE network. This approach removes A further uncertainty is due to two promising – and to some extent the controversial coexistence of LTE and Wi-Fi in the unlicensed 5 GHz band, complementary – alternatives: MulteFire and LWA. but it also removes the LTE performance advantage.

MulteFire allows for the use of LTE as the air interface in the 5 GHz band, As long as devices can support LAA, MulteFire and LWA (and they likely will), without needing to use a licensed band for the control plane. So venue mobile operators will have multiple ways to use unlicensed spectrum in owners or operators that do not have an LTE network in licensed bands can addition to Wi-Fi – as long as they get access to the venues they want to deploy MulteFire and, if they choose, offer LTE unlicensed access on a cover. We expect operators to select one or more of these solutions, as neutral-host basis to mobile operators. This approach enables operators to soon as they deploy small cells, because using unlicensed spectrum not only improve capacity and/or coverage in venues where they may not have augments capacity, it significantly improves the business case for small cells, access or where they do not want to deploy licensed LTE. which to date has been a difficult one.

Massive densification requires new spectrum to manage the increase in traffic Millimeter-wave bands are the other hot prospect for spectrum expansion. The loads and meet performance requirements efficiently and cost effectively. By amount of spectrum available is huge and, because of the high reuse that high packing the infrastructure closer and using multiple layers (e.g., small cells and frequency makes possible, the potential increase in capacity is astounding. macro cells), operators can reuse a frequency channel more intensely. But at some Spectrum in these bands is going to be much less expensive, or even usable on an point they face a diminishing return: the marginal investment to increase capacity unlicensed or lightly licensed basis, and current 5G standardization efforts cover becomes too high for the increase in performance it brings. mmW bands both for fixed wireless links (e.g., backhaul or possibly fronthaul) and for access in areas of very high density (e.g., where even small cells or DAS are not Many operators face this situation in the macro RAN. They increase the density of sufficient, the extremely hot spots). MmW supports densification on two fronts – base stations, added MIMO and CA, and split sectors, and they reach a plateau. xhaul to small cells, and access from UEs – which can be combined when using in- The next step is to add small cells in the same band, but that introduces band xhaul. interference – especially if the small cells are located outdoors in areas covered by the macro layer. In some environments the interference can be managed Using mmW for access requires more than new antennas or distributed small cells. effectively, but it adds a cost in terms of effort and lost spectral efficiency. Because of the small coverage radius, mmW access can generate large numbers of handoffs, as users will cross the cell-edge border frequently and move from mmW Using multiple bands to cover the same location enables operators to lower the to sub-6 GHz cellular coverage from macro or small cells. Frequent handoffs create impact of interference and minimize costs. Low-frequency bands are still valuable high levels of overhead that affects the anchor sub-6 GHz network. for improving coverage in low-density areas and for some IoT applications. In high- density environments, higher-frequency bands are more effective in reducing the A solution to this problem is to implement a phantom-cell architecture, in which interference and increasing spectrum utilization, because of the more limited coexisting elements – e.g., sub-6 GHz and mmW – become part of the same cell ID coverage range. This is a major limitation in a macro network, and for this reason and the control plane is managed for both bands from the sub-6 GHz bands, which mobile operators have been reluctant to use spectrum above 2.5 GHz to date. have wider and more consistent coverage. Load balancing across bands enables operators to direct traffic to the band that can accommodate it more efficiently. Increasingly, however, mobile operators have become keen to use higher frequencies for lower layers. Candidates range from the 3.5 GHz band to the Further densification and more intensive spectrum utilization may come from unlicensed 5 GHz band, and all the way to mmW. The 3.5 GHz band is an excellent device-to-device communications – either subscribers or IoT devices. Device-to- candidate for small-cell deployments, because its shorter range strikes a good device communications can establish an ad hoc, mesh-like network that expands coverage/capacity balance. However, the amount of spectrum available is limited, the reach and capacity of the rest of the network. This possibility calls into and in some countries there are regulatory restrictions or spectrum allocation question the sharp boundary between network and device that is prevalent in issues that have been slowing down the plans to use the 3.5 GHz band. today’s networks.

Initial densification efforts with small cells and DAS have used a co-channel model result, operators have been reluctant to deploy small cells widely, and have in which they share spectrum with the macro layer, but generate different levels focused on areas where the need for additional capacity makes them less of interference. Femtocells, the first incarnation of small cells, were largely sensitive to cost. deployed indoors, at very low power and in places where macro coverage was weak or absent, and so interference was not a big issue. Two ways to cope with the interference issues have emerged and are likely to accelerate massive densification. The first of these involves spectrum bands that Outdoor small cells were initially deployed in many cases to address areas are not traditionally used for mobile and that, therefore, are less expensive and in without coverage. Interference was, by definition, not an issue. When small-cell higher-frequency bands – such as the 3.5 GHz, the unlicensed 5 GHz band and deployments moved to dense areas where operators had macro coverage but mmW bands. Operators can use these bands in addition to co-channel spectrum insufficient capacity, interference became a major issue that slowed the rollout of within the same small-cell enclosure, or they can deploy a sublayer in these bands small cells. It also pushed operators to spend more money on increasing macro and let the macro layer retain control of cellular spectrum. In both cases, the cost capacity, and to move to small-cell deployments as a last resort after all the per bit decreases. In the first case the low incremental cost of adding these bands, macro enhancements had been used. Initially, techniques like CoMP or eICIC coupled with the increase in capacity, reduces the per-bit capex and opex. In the were not sufficiently mature or widely deployed, making interference second case the business case is strengthened by preserving capacity in the management less efficient. But more fundamentally, interference between small macro layer by avoiding interference. and macro layers in co-channel deployments reduces the capacity of the macro layer, and that is highly undesirable and expensive for mobile operators. Even if A second way of coping with interference is to move the lower-layer the cost of installing a small cell is lower on a per-bit basis than a macro cell, the infrastructure indoors. In locations where per-small-cell deployment and cost savings could easily dissipate if the small-cell installation results in a operating costs are comparable or lower than in an outdoor environment, indoor reduction of the macro-layer capacity. networks can help reducing the per-bit capex and opex. The reduced impact from interference lowers the per-bit cost in indoor deployments. However, in-building Of course, an effective way to minimize interference is to move away from co- infrastructure faces business model, ownership, and control challenges that are channel deployments and use a separate channel. Operators largely resist this, different from outdoor infrastructure, and access to indoor locations may not be because typically cellular spectrum is too expensive to use only in the small-cell available or affordable to mobile operators. The move to in-building layer. In most environments, a co-channel deployment supports a more efficient infrastructure will undoubtedly intensify and accelerate the densification process, use of spectrum, in terms of capacity density (bits per sq km), than a deployment but it will not eliminate the need for outdoor infrastructure, which will still be in which small cells use a channel different from that used by the macro layer. But used to meet demand from outdoor subscribers and in places where operators the reduction of capacity from interference makes the business case for small cannot deploy indoor infrastructure. cells less palatable, because it increases the overall network per-bit cost. As a

A vexed question in the small-cell ecosystem is whether backhaul should be wireline (and, if so, whether it has to be fiber) or could be wireless (and, if so, in which bands). The debate does not regard indoor deployments, where wireline backhaul is typically used within the building. In outdoor environments, however, both wireline and wireless solutions present benefits and limitations – and approaching backhaul as a combination of fiber and wireless links may help reduce the drawbacks of both technologies.

Fiber is the ideal backhaul for outdoor small cells, but it is not always available, and even when available, it is not always cost effective. In most environments, fiber is available in the vicinity of the small cells, but bringing it to the small cell is often too expensive, not to mention time consuming.

Wireless backhaul is the alternative, as long as small cells do not use a C-RAN or virtualized architecture. With remote baseband, fronthaul requirements are tighter and wireless fronthaul is possible, but it requires specialized solutions. Solutions that work for backhaul typically do not have enough capacity or a low enough latency to support fronthaul. The limitation of wireless backhaul is that to provide sufficient latency, line of sight from the small cell to the aggregation point is required. As the link length grows, the likelihood of having a reliable line of sight and reliable performance decreases. Multi-hop wireless backhaul can compensate for the lack of line of sight, but increases costs and latency.

In outdoor environments, fiber and wireless are not mutually exclusive, but rather two components defining the backhaul. In fact, a mix of fiber and wireless backhaul is the dominant solution, with the choice between the two dictated by cost and availability tradeoffs. The backhaul for every small cell terminates into fiber; what changes is the length of the wireless backhaul, which can be zero for a small cell with fiber backhaul.

Another concept that pervasive networks challenge is the dichotomy of backhaul and fronthaul in terms of requirements and what technologies could meet them, and in terms of what type of signal the fronthaul carries.

Within mobile networks, fronthaul is the link from the RRH to the BBU, and backhaul the link from the BBU to the core network. If the RRH and BBU are co-located, there is no need for fronthaul. Because the fronthaul carries the analog signal, the bandwidth and latency requirements are much higher than those for backhaul. As a result, some technologies and solutions may be suitable only for backhaul. Others may be well-suited for fronthaul, but too expensive for backhaul.

First off, with the move to extra-low latency and high capacity (e.g., when using mmW for access), the 5G backhaul requirements may approach those for fronthaul today. But this also means that fronthaul requirements, if using CPRI, will also grow to alarming levels. This creates the possibility that the fronthaul may become the bottleneck, and the risk that RAN capacity may have to be capped if the BBUs are remote. (Another possibility is that RAN virtualization will slow down because fronthaul requirements are too onerous.)

Because it is difficult to meet current and future fronthaul requirements with CPRI, especially in small-cell deployments where dark fiber is too expensive or not available, there is considerable ongoing work to define alternative interfaces – e.g., compressed CPRI or Ethernet – and functional splits, in which some of the baseband functionality stays at the edge, co-located with the RRH.

With functional splits, the fronthaul requirements decrease, but so do the benefits of virtualization. Multiple options are available, and operators need to decide what level of functional split works best in each location in their networks. It is not yet

REPORT Massively densified networks © 2016 Senza Fili www.senzafiliconsulting.com |23| clear whether there will be a dominating functional split or which splits will be best suited to which environments.

Importantly, functional splits call into question a clear-cut distinction between fronthaul and backhaul, and instead suggest a continuum of functional splits between RRH and BBU, which recently has been referred to as the xhaul. Then, depending on the type of xhaul – i.e., the selected functional split – and the RAN requirements, different interfaces and solutions become appropriate. The xhaul approach recognizes operators’ need for flexibility as they densify their infrastructure and use a wide range of RAN solutions to strike the right balance between costs and performance.

When planning for densification, operators need to select not only technologies, solutions, spectrum bands and locations, but also the topology of the densification. There, access point density is a choice that is becoming more prominent, although it is rarely discussed as a stand-alone dimension of the densification strategy. As a wider range of solutions, spectrum bands and business models become available, operators can choose among options that vary in power, coverage, cost, equipment size, and ability to fit into multi-operator, neutral-host models.

Given a target capacity they aim to have in a given location, operators can choose to have a high number of low-power, reduced-coverage, low-capacity access points, or a smaller number of high-capacity access points. When using mmW for access, or multiple bands in the same access point, operators can create super hotspots where they have a very high concentration of traffic.

The primary consideration in selecting the appropriate access-point density is the distribution of traffic. Subscribers and the traffic they generate are distributed very unevenly, so it is crucial to place the access points as close as possible to subscribers. If the distribution is uneven and highly clustered, access points will be placed more densely in high-usage locations. If the distribution is more even, access points with larger coverage areas may be preferred.

In addition to traffic distribution across locations, operators need to consider backhaul availability, deployment costs, spectrum availability (including traffic load in unlicensed bands), business model (e.g., whether there is infrastructure sharing among operators or a neutral-host model) and RF propagation in the environment. For instance, if deploying access points is inexpensive, backhaul is easily available, or spectrum availability is limited, an operator may opt for a denser network of access points with limited coverage and capacity. Fewer but more powerful access points may be better suited in locations where installation and backhaul are expensive.

Densification demands that the wireless infrastructure move closer to revenues or to get better mobile service or both, but they do not know how to subscribers, vehicles, and IoT sensors and other devices – and this means structure deals with operators because it is uncharted territory. The same is beyond the cell-tower model, which continues to be used for the macro layer true for mobile operators. As a result, we see trials, negotiations and even legal but needs new indoor and outdoor models as complements. challenges – but little in the way of the expected large deployments. The urgency of densification is increasingly felt by both parties as bad wireless In a macro environment, the operator has full control over network planning, service affects operators and venue owners or administrators, and so deals are deployment and operations, using telco assets such as cell towers. The starting to come together. They will undoubtedly evolve through time as we operator owns the end-to-end infrastructure and manages traffic and understand the deployment models better from technical and business interference. Cell towers are usually owned by third parties, but operators lease perspectives, but these early deals are the necessary first step to get beyond ad space on them and retain control of the telecom equipment. This model, well hoc densification to large-scale densification. understood by every player in the ecosystem, ensures that deployments proceed smoothly. More interestingly, the need for mobile operators and asset owners to work together – directly, or indirectly through third parties such as service providers, In a sublayer deployment, this model no longer works. The access fiber providers or cell-tower companies – is also accelerating the development infrastructure has to be mounted on non-telecom assets, such as lampposts, of closer relationships with cities, enterprises and other venue owners and exterior or interior building walls, ceilings, advertisement displays, and public institutions. These relationships can lead to better performance and to the transportation vehicles, or even below street level. This creates constraints on creation of services that mobile operators can develop or support. In a where operators can deploy the equipment; it also creates the need to densified, pervasive network, it is not just the equipment that moves closer to establish direct or indirect relationships with the owners of these assets – cities subscribers – it is the relationship among subscribers, venue and public entities, enterprises, educational institutions and other venue owners/administrators, and operators that becomes tighter and deeper. owners. In some cases, these entities expect to extract rent from mobile operators or third parties working on their behalf. In other cases, the asset The need to negotiate deals with asset owners coupled with the need to find owner may pay for the infrastructure or encourage rent-free installation and cost-effective ways to deploy small cells drives new business (and deployment) operation, but may require visibility or some level of control over the local models. The initial small-cell business model assumed each operator would network. deploy its equipment independently of other operators or service providers, striking deals with asset owners. But that is too expensive, time consuming and Negotiating deals with the new asset owners has proven to be one of the inefficient to scale beyond prime locations (e.g., some parts of Manhattan, biggest challenges – and causes of delay in deployments – that operators face. downtown San Francisco, central London) to areas with less concentrated, but Asset owners are eager to host the telecom infrastructure, either to extract still high, traffic.

REPORT Massively densified networks © 2016 Senza Fili www.senzafiliconsulting.com |26| Especially in indoor venues, the trend is toward shared deployments, in which This model does not exclude opportunities for the operators and venue owners one operator or, more commonly, a neutral host acts as the interface between to negotiate specific arrangements about network performance, coverage or the asset owner and the operators that wish to participate in the local network. functionality, or to provide additional services (e.g., enterprise services, or IoT services to city agencies). These arrangements will encourage the venue This is very similar to the DAS neutral-host model, in which multiple operators owners to participate more actively in the densification process – in part by can share the DAS, but each controls its own transmissions. This model enables funding deployments, and in part by requesting services – and to see the operators to reduce costs by sharing some of the infrastructure, and it wireless infrastructure as an integral component of the venue, comparable to streamlines deployment and operations, because the neutral host manages all the in-building electrical network. the relationships with venue owners on one end, and with multiple operators on the other end. There is a widespread belief that the DAS neutral-host model In this context, the availability of the 3.5 GHz band in the US and of other sub-6 does not work in small-cell deployments because it requires operators to share GHz bands in other countries will push the shared deployment business model the RAN – an option that nearly all operators consider unacceptable if it further, because it allows neutral-host players to install a small-cell network in a involves their licensed spectrum. But the neutral-host model can equally allow venue, using spectrum bands that are not allocated to a specific operator. The operators to deploy their own radios, retain control of the use of the spectrum venue owner or the neutral host can deploy the network and allow access to and manage traffic, without having to negotiate a separate deal with the asset operators on a wholesale basis. A benefit of this approach is that the neutral owners, and without having to directly deploy and manage the RAN host or venue owner can deploy a single network that can be shared, and this equipment. results in more-efficient spectrum utilization and in lower costs.

Many mobile operators worldwide question their future relevance – and worry about their ability to extract the revenues they deserve from their networks. Typically, these concerns are rooted in the risk they perceive that they will become a dumb pipe that can transmit an increasing amount of data to their subscribers and do so more reliably than in the past, only to see subscribers value the networks less than the OTT apps they use on those networks.

The transition to massively densified, pervasive networks can change this. The role of the mobile operator is transformed by the increased complexity of wireless networks that have to optimize and integrate multiple spectrum bands, technologies, services, device types and topologies. Mobile operators can no longer focus only on pushing as many bit/s as their infrastructure supports; they also need to manage traffic, applications and network resources in a much smarter way than they are accustomed to. Increasingly, they do not look like a utility – they look instead more like orchestra conductors, coordinating transmission in a multi-layer network – a network in which they have the flexibility to set strategy in ways that differentiate their network from that of their competitors.

II. Vendor profiles and interviews

REPORT Massively densified networks © 2016 Senza Fili www.senzafiliconsulting.com |29| Tools helps mobile operators with documentation Anritsu solutions also help operators with Profile and reports, real-time analytics, automated densified networks to identify the different assessment of RF sweeps and PIM test results. sources of interference that affect macro-cell and Anritsu small-cell networks and to manage interference, if The ability to automate and scale testing and necessary, in real time. monitoring in wireless networks is crucial for Founded over 120 years ago, Anritsu Corporation operators moving to multi-layer, multi-RAT The portfolio of Anritsu measuring instruments is is a global provider of communications test and networks and with DAS; the number of tests well suited for indoor densified networks. It measurement solutions. needed to assess performance rapidly increases consists of solutions for both the wireless and with complexity and makes manual field testing optical segments; they can test and monitor both Anritsu’s measuring instruments support multiple time consuming and expensive in terms of staff the access and backhaul/fronthaul portions of areas: resources. mobile networks.

Mobile wireless communications, RF and microwave: base station analyzer, Bluetooth and WLAN tester, cable and antenna analyzer, conformance test system, interference hunt, PIM analyzer, peripheral equipment, power meters and sensors, signal analyzer, spectrum analyzer, signal generator, signaling tester, shield box, trace management, vector network analyzer, handheld vector network analyzer Digital broadcast: Digital broadcast analyzer Devices and components test: Bit-error-rate testing, eye pattern analyzer, vector network analyzer, signal generator, optical spectrum analyzer, peripheral equipment Transport: IP/Ethernet testers, SDH/SONET/OTN analyzers, PDH/DSn analyzers, multi-layer network test platform Optical: OTDRs, multi-layer network test platform, optical loss test set / light source / optical spectrum analyzer, video inspection probe In addition, Anritsu has recently introduced SkyBridge Tools™ to manage cloud data. SkyBridge

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |30| With LTE in particular, a clean uplink really helps Monica: Is automation going to help this? Anritsu capacity, and capacity is what densification is all Networks are becoming more complex, not less about. complex. How is testing going to evolve with the Managing increase in complexity? Monica: You’re personally more involved in the DAS testing, but at Anritsu, you cover other areas Tom: Automation can help this. In manufacturing, interference with respect to densification, too. for many years, there’s been a piece of software called a test executive. It sets up tests, it in densified networks Tom: Personally, I’ve been involved in DAS testing automates the tests, it runs the tests, it collects the A conversation with Tom Elliott, and interference. Others in Anritsu are involved in results, and generates statistics. the full range of tests that we do, as well as Sr. Product Manager, spectrum monitoring. We need something similar to that for the field, Anritsu Company something to automate the field tests, and DAS is Monica: What is different about DAS testing? The a perfect setup to do this. At the same time, end result is to make sure that the QoE is good for there’s the idea of removing some of the Monica Paolini: In this conversation with Tom the subscriber. possibilities for error out of the DAS testing. Elliott, a Senior Product Manager at Anritsu in the United States, we discuss how testing evolves as Tom: In a tower installation test, you have cables In the work we’ve done with some of our DAS operators densify their networks with DAS, small going up the tower. You have antennas, maybe installers, we’re finding when we come in, there cells and other solutions. some splitters, and other passive RF components. might be a 10% error rate on these files. Say I have A typical tower test would have 25, 50, 100, 5,000 files, I sample 500 files. I can’t check every Tom, could you give us an introduction on the maybe even 150 sweeps of some nature. We’re one of them, so I sample. approach to densification at Anritsu and what your talking about return loss, distance-to-fault, cable role is within the company? loss, PIM, and maybe some fiber test. That can be I say, “OK, it’s good.” I pass them off to my end handled by the existing processes. customer. They look at 5,000 tests and say, “What Tom Elliott: As far as personal work on am I going to do with this?” They sample the tests. densification, I’ve been heavily involved with When you move to a DAS system, especially a They sample different tests, and if there’s a 10% redefining what field testing means for DAS. The neutral-host DAS system, where you need to test error rate, chances are they’re going to find some idea of a test as being a return loss, or a distance- three or four frequency bands on every cable, problems. The results all come back to the testing to-fault really isn’t working too well for the DAS those tests multiply, and they multiply contractor, and we get into this loop. people. dramatically. The problem was expressed to me best this way: “I We’ve also had a focus on received signal quality. I A medium-sized DAS install could have several have 1,000 traces in a directory on my PC. Which and others have been heavily focused on the idea thousand tests. A football stadium, for instance, ones are missing? Which traces are duplicated? of spectrum assurance, the idea that you can have may have as many as 15,000 tests. It’s the sheer Which traces are misnamed? I haven’t even gotten a variety of ways to make sure that your uplink is scale that becomes a problem. You’re talking to which traces fail judgment.” The existing clean. about months, man-months, spent dealing with processes, which are based on a tower technique, these tests. just do not scale.

Tom: One thing Anritsu is doing is redefining what a test is. I’ve talked about a return-loss sweep, a distance-to-fault sweep, a cable-loss sweep, and about having to do these all at three different frequencies.

Why do we need to run nine lab-based tests on each cable and deal with these individually? Don’t we have computers? Don’t they do things like this? We do have computers, and they do things like go back and retest.” These sorts of things could be before they’ve closed out, before they’ve gone this. We can redefine what a test is, from “Here’re removed by test automation. somewhere else, before they have to come back, all these things I have to do to this cable,” to “This before payment is delayed. cable is good or bad.” We can remove some of Monica: In a DAS environment, you might have a these workflow issues. neutral host, so it’s not necessarily the operator Monica: Basically, you can see there is a problem, doing the testing. And the people that are doing and then you can dig down specifically in that area, There’s a MOP, a Method of Procedure. Some the install and the testing might not be as whatever the case may be. companies call this a SOW, or statement of work. qualified, or have so much RF experience. Doesn’t By necessity, it’s fairly general, but when a that create additional challenges? Tom: That’s exactly it. The idea is to get things technician is faced with an ambiguous situation, done quickly and done now. Our field tests show what they’ll tend to do, rather than have to come Tom: It certainly does, and I’ve been involved in that SkyBridge can cut the actual test time by up to back and test again, is they’ll take three, four, or some of these. DAS testing, even if you do have RF 90%. That’s a big number, but if we’re going from a five different traces, and let the engineers sort it experience, is a different field than some of the manual method to a computer-aided method, out. other testing you may have done, and there are that’s not surprising. certain ambiguities there. Typing filenames on the instrument turns out to Monica: With densification, the more packed your be a major time sink. Not only do they tend to What Anritsu has proposed is an automated field infrastructure is, the more opportunities you have make a typo once in a while – you would, too, if it test solution. You may have heard of it by the for interference. That is whether it’s indoors or was 20 degrees out and blowing like anything – name of SkyBridge Tools. It’s on our external outdoors. How do you deal with that? but each technician tends to have his own naming website. It provides setup information for the standard, and they leave it to the engineer to sort instruments. It provides test automation when Tom: First, maybe I should take a moment and it out. you’re on site, and it provides automatic define what the interference mechanism is. judgment, automatic reporting with cloud-based There’s also some ambiguity in the existing dashboards, so everybody with a login knows Interference is a receive issue. Signals that get to process for instrument setups, and this can lead to exactly how the job is progressing. the radio’s receiver affect the front end, and even situations like “Oh my goodness, it was the wrong if it’s not the signal you want, it will come in there. start and stop frequency,” or “The limit line was This way, the questions can be asked and It will reduce the sensitivity of your radio receiver. wrong,” or “Something was wrong. We’ve got to answered while the technician is still on site, This lowers the radio’s sensitivity, and increases

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |32| your bit error rate, increases your frame rate, because it’s above the small cell’s noise floor. We signal or a strong transmitter will overwhelm a increases your dropped calls and all that bad stuff. can coexist perfectly happily. weak desired signal.

The name for this is receiver desense, or If we’re 50 ft from a small cell, the signal will lose All of these are typical interference sources, and desensitization. If it’s severe, it’s called receiver 56 dB over 50 ft. That puts the signal at -96 dBm, we’ve got a great app note on interference blocking. at the small cell’s receive and it’s a problem for the hunting concepts that goes over this in more small cell. detail. One thing that seems to be a chronic misunderstanding about interference is that My point here is that when you put in small cells, As far as making interference hunting more interference does not have to be on your receive there’re going to be interference sources that will efficient, Anritsu does have a tool set for this, channel. All it has to do is get to the input of your bother a small cell that a macro cell would never starting with some of the very simplest traditional receiver. It can be on a different frequency. see. Interference becomes a bigger problem. Of methods. course, the small-cell reception area is smaller. It’s In other words, it can be in band, but it doesn’t a small cell, after all. But there are other We’ve got the traditional direction-finding tools. have to be on channel, so your receive pre-filter interference sources that will matter to the small We take a spectrum analyzer and a Yagi antenna, matters. If it gets through your receive pre-filter, if cell that a macro tower won’t see. Efficiency in look for the strongest signal, and you triangulate. it gets to the input of your receiver, it’s finding interference becomes very important. We have map-assisted tools that will actually put a interference. map on a spectrum analyzer. We have car-based Monica: How do interference sources change as signal location, where we can go through and Monica: In terms of the testing, how can you have you move from macro to small cells? essentially seek the power. your testing become more efficient where you have small cells, which are additional sources of Tom: It depends on the sort of sources we’re Do you remember the child’s game of hot and interference, when small cells and macro cells are looking for. There’s on-channel interference, cold, where one child picks something, and the in co-channel deployment? there’s interference that’s off channel but in band, other child asks, “Am I getting hot? Am I getting there’s impulse noise, arcing, sparking. There are cold?” It sorts of works that way, and it’s Tom: Anritsu has a lot of interference solutions, even still jammers around. surprisingly fast. Also, it’s reliable, because it takes but there’s one thing I need to point out, which is care of issues with multipath, echoes, reflections, that the incoming signal amplitude at the receiver We had a case a while ago where a high school and even diffraction. matters. teacher had a jammer running during his tests so his students couldn’t cheat, and it was shutting Another solution is monitoring, because signals Small cells have their own unique problems when down an AT&T sector. Jamming 911 calls? Not aren’t always interfering. We can characterize a it comes to interference. good. signal by doing short-term monitoring. Every one of our spectrum analyzers is web enabled. You can Let’s say I have a signal source, an interference We also have harmonics, multiples of an original hook them up to the internet. You can control source. It’s at -40 dBm. It’s emitting at -40 dBm. If signal. Some of our TV signal harmonics fall in the them with a browser from a distance. We can drop that signal is a mile from a tower, propagation PCS band. There is intermodulation, both active something off at a site for a week, for two weeks, models say it’s going to lose 96 dB by the time it and passive intermodulation, that we all know for three weeks, and see what’s happening. gets to the tower. In absolute numbers, It’s going about from PIM testing. There’s something called to be -136 dBm at the tower. It’s not going to be the near-far problem, where a strong interference On the other hand, you might be interested in an issue. The tower won’t see it. It won’t matter some long-term monitoring. We have a set of

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |33| headless spectrum analyzers designed specifically That’s not possible with LTE. The bands don’t Tom: Real time can be a day, it can be an hour, it for this sort of task, which can be either permit it. If you want to have more capacity, first can be a minute. Picture this scenario. There is an temporarily or permanently in place, and we have you need to make the most out of the macro cells engineer at a central site somewhere in the world. software to go with it, so you can maintain your you have. That’s what spectrum monitoring is There’s a technician on site in the field. The signal quality throughout your network as things about. technician runs some diagnostics. He posts the change. The device is comfortably monitored results in the cloud. The engineer sees the results remotely. Second, once you get your small cells in, or your and downloads some more tests to the iDAS or your oDAS, you need to make sure it’s technician’s equipment. That sort of collaboration Monica: There is testing that you do at the performing effectively. is what I am talking about. beginning, but then monitoring is an ongoing effort. Is the relationship between testing and Monica: What is the impact that you’re seeing Monica: I guess at different sites, and for different monitoring evolving somehow, with densification from densification on spectrum assurance? needs, you might switch to different time and networks that are becoming more complex granularity while there? and dynamic? Tom: Spectrum assurance is Anritsu’s name for a family of spectrum monitoring, interference Tom: Absolutely. Monitoring typically uses a Tom: Back when we were mostly concerned about hunting, and signal mapping tools, both indoors 15-minute window. Troubleshooting would coverage, interference didn’t matter so much, and outdoors. We’re developing these in response require updates in a few seconds. If we’re building because all of our cellular standards have to our customers’ requirements for exactly the a DAS system, maybe once a day is enough. It redundancy, and they can handle a certain amount sort of capacity we’ve been talking about with LTE depends on the task. of interference without dropping the call. Every and densification. one of those methods of handling interference Monica: Monitoring allows you to identify and fix a involves sending out more bits to get the same Monica: Where do you see testing moving in the problem, but also it allows operators to optimize amount of data. In other words, you’ve affected future? the use of the network resources. Can you help your capacity. them with that as well? Tom: We’re going to be moving closer to real time. As soon as we moved to LTE and we were We have all this communication capability now. Tom: Yes. We do have a product line that provides concerned about capacity and densification Our instruments are using it and they will continue a big-iron software application, called Master suddenly, interference matters, and it matters a to use it. Claw, that is a service-assurance solution. lot, because it’s hitting your throughput directly. That’s why monitoring is becoming of interest to Cloud-based solutions are tremendously efficient. We also have a software application called Vision. our customers. Network operators, especially, are The test automation for DAS is a cloud-based It works with our long-term RF monitoring probes, concerned about their signal quality and their solution. We’re going to see that. You’re going to and it helps characterize when signals occur and throughput. see more control, you’re going to see more what they look like; it helps figure out where the remote expertise, you’re even going to see remote signals are occurring, within a few blocks, by If I go back 10 years to CDMA, if we had any dispatch coming in the future. triangulation; and it provides all the necessary interference problem, it would be the easiest thing information to dispatch a team to go and find that just to plunk a $4,000 board into a base station Monica: You mentioned real time. What is real signal. and bring up a new carrier, and all of a sudden, time for you in testing? Is it a day, an hour, a you have more capacity. minute, a millisecond? Now we can sit in a central place somewhere, let’s say somewhere in the US, characterize an

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |34| interference on three or four sectors in, say, Los that virtualization? How is virtualization in the We expect this sort of collaboration in field tests to Angeles, and dispatch a local team to go find it cloud changing your solutions? continue. After all, field tests naturally have when we’re ready to. That’s the sort of real-time, experts in one location, and the man on the continuous, cloud-connected thing we’re talking Tom: Obviously, our solutions are becoming more ground in a different location. As soon as you do about here. connected, and you can expect that to continue in that, electronic communication becomes just a the future. We have a wide world of connection, at natural solution. Monica: Real-time optimization, and then you also least electronically. Barriers are falling, as mentioned the cloud virtualization part. What is evidenced by this on-line interview. We’re many miles apart, and it’s working just fine.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |35| About Anritsu Anritsu Company is the United States subsidiary of Anritsu Corporation, a global provider of innovative communications test and measurement solutions for 120 years with offices throughout the world. Anritsu’s “2020 VISION” philosophy engages customers as true partners to help develop wireless, optical, microwave/RF, and digital instruments, as well as operation support systems for R&D, manufacturing, installation, and maintenance applications. In addition to supporting precision microwave/RF components, optical devices, and high-speed electrical devices for communication products and systems, Anritsu provides a large portfolio of solutions to meet the growing demand for in-building wireless services from DAS to small cell environments. About Tom Elliott Tom has 20 years of experience in the telecomm industry working with RF and cellular technologies. Tom has concentrated on test and measurement for cellular base stations for much of this time. Tom is a Product Manager for Anritsu Company with worldwide responsibility for wireless service providers. He focuses on improving network performance and making technicians more productive through the technologies and tools of Anritsu. Tom has taught hundreds of technicians, written several procedures and courses, and regularly receives the input of technicians, managers, directors, and CTOs on new test requirements as the wireless network evolves.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |36| predicted, and can resolve the performance issues Other solutions in the TEMS portfolio that focus on Profile they identify. the RAN include:

Ascom Network TEMS iBuildNet enables engineers to detect TEMS™ Automatic, a tool for automated data discrepancies between the real environment collection for testing, monitoring, Testing versus the plan, using the latest site survey benchmarking, audit and verification, via information provided by TEMS iBuildNet Walk. probes that can be remotely controlled from a

With this solution, operators can collect data, and central location. add text comments, video, photos and 3D models, TEMS™ Investigation, for drive testing of the For more than 25 years, Ascom has helped mobile as they deploy and test the networks. air interface and service quality. operators worldwide with testing, monitoring and TEMS™ MobileInsight, to test QoS from optimizing solutions that span the entire wireless TEMS™ Discovery also leverages data collected by mobile devices, using QoS-specific KPIs, SLA infrastructure – from the core all the way to the TEMS Pocket or TEMS Pocket Remote to identify monitoring, crowd sourcing, and subscriber RAN. RF problems in the network, and geolocate their feedback. source. It allows for analytics and reporting in TEMS™ Monitor Master, to test and monitor In the RAN, densification has played an important indoor testing, drive testing, and network network performance using simulated traffic role in shaping Ascom products over the years, but diagnostics. at the application layer for different devices. with the introduction of IP-based multi-RAT, multi- TEMS™ Symphony, for benchmarking. band and multi-layer networks, Ascom has developed solutions specifically targeted at operators deploying small cells, DAS and, more generally, HetNet architectures.

Ascom’s TEMS™ iBuildNet is a planning and design tool for HetNets which allows operators to design deployments that combine both indoor and outdoor infrastructure. It models multiple elements that affect performance, such as coverage, traffic steering, path loss, signal level, SINR, cell overlap areas, best-server maps, and LTE and Wi-Fi handover – in 3D if desired – to determine the required coverage in indoor locations. According to Ascom, TEMS iBuildNet can cut planning and design time by 50%.

With TEMS™ Pocket and TEMS™ Pocket Remote, operators can measure indoor and pedestrian area traffic from handheld mobile devices. Specifically, these products can identify locations where performance differs from what the initial plan

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |37| But before we get to this, we should define what Video’s going to continue to be a huge driver. Ascom densification is. Everybody’s talking about it, but What it boils down to, Monica, is quality of are we all talking about the same thing? experience. This is the whole purpose for a carrier Testing and to go into a partnership with an enterprise or a Todd: The way I look at densification is as a means commercial segment. And a hotel wants to by which a carrier goes about improving the improve its customers’ experience, and part of monitoring in capacity of its networks, especially in densely that experience is, of course, a mobile data populated areas and large public venues. connection, whether you come indoors or go densified networks outdoors. A conversation with Todd Cotts, A lot of the capacity issues are driven by the demand of mobile users, especially data demand Carriers are focused on quality of experience. It’s Product Marketing Manager, these days. As the data demand increases, the been shown that the way customers evaluate their Ascom Network Testing load is greater on the network. quality of experience is by speed. That’s one of the ways. Another one is video quality. We have all The capacity that operators originally planned the experienced stalling during the video. Monica Paolini: My guest in today’s conversation network to support is no longer sufficient. The is Todd Cotts, the Product Marketing Manager at traffic load is now straining the network. What Another way to define quality of experience is the Ascom Network Testing. happens is that the coverage begins to shrink time required to download a web page on a when capacity is insufficient. People and calls are mobile device. When the network is slow, it’s very Todd, can you tell us what Ascom is doing to dropped off. The speed slows down. Latency frustrating. support densification efforts? becomes an issue. Then, you have the resolution of the video Todd Cotts: Ascom Network Testing has been That’s really what densification is all about. And content, as well as the responsiveness of a mobile around for several years. What we’re known for at the way carriers are going about doing this is app. All of these things amount to one big, key Ascom Network Testing is our TEMS solutions: our through small-cell deployments. driver, which is quality of experience. TEMS products and TEMS services. For over a quarter of a century, we have been providing Monica: How are operators addressing the If you don’t get the quality of the experience right, products and services in the network testing field capacity problem? Is the strategy the same across whether you’re a carrier or an enterprise or to the telecommunications industry, including all operators? commercial segment, whatever it is you’re doing, of the top, tier-one US operators. you’re going to churn. Todd: The drivers are the same. Mobile-data traffic Many of those services have been around the has increased exponentially and will continue to Tenants are going to leave your buildings. Hotel network testing area, but recently – earlier this do so into 2020. Much of that mobile data traffic, guests are not going to return. It’s a big thing for year – we’ve launched a new solution that focuses 80% of it, is coming from indoors. A lot of it is from everybody, and everybody has a stake in it, not on the design and deployment of small-cell and the commercial and enterprise segments. just the carriers. heterogeneous networks. When it gets down to the drivers, there are a few There’s also a near-future piece to this, if I may. Monica: With densification, there are many new things I want to iterate here. In 2015, for example, We all talk about 2020. One of the near-future challenges, but also wider opportunities for 15% of mobile data traffic was from social media, pieces of this is that, as I mentioned, video testing, because the role of testing is evolving. but, larger than that, 50% of it was from video. outstrips mobile data traffic, and it’s going to

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |38| continue to grow tenfold by 2020. In fact, there’s research that says about three-fourths of the mobile data traffic by 2020 is going to be video. That’s globally.

There are some other challenges, we’ve got to realize, in the next four years. We don’t have a lot of time to do what we need to do – improve quality of experience, expand the network, and improve capacity.

Around the corner, you’ve got new mobile apps and services. You’ve got the internet of things. You also have smart cities. Smart cities are popping up everywhere. In the San Diego area, for example, a completely new urban development has popped out, and it’s all smart-city based, where all the digital devices are connected and talking to the internet, talking to one another.

Then, we have the dense urbanization itself. As the urban populations continue to explode, and as Todd: Gartner’s predicting, by 2020, 25 billion which are really hybrid network solutions that smartphones and smart devices continue to be connected devices. That’s a lot of connected combine multiple RAN technologies as well as adopted, and as Wi-Fi and 4G continue to be used, devices. multiple access node types, such as iDAS and you have this huge explosion of mobile data traffic oDAS. demand on the network. Monica: Densification is often narrowly thought of as centered around small cells, but there are Wi-Fi’s a part of that, too. Not just enterprise If we’re not prepared to support that, through different ways to go about it – and different Wi-Fi; we’re also talking about carrier Wi-Fi these small-cell deployments, Wi-Fi offload, carrier Wi-Fi technology solutions. Initially it was just small cells days. And all of that is part of a small-cell or other technologies, the quality of experience is on lamp posts. Now, it’s much more than that. deployment, depending on the need that’s been going to go downhill. What are you hearing when you talk to operators? identified by the carrier.

Monica: With IoT, we’re moving beyond Todd: When you’re talking about densification, Monica: The networks are becoming more subscribers’ quality of experience. You typically do you’re really talking about small-cell deployments. complex. In testing and monitoring, your work is not have a human at the other end, but you do But a small-cell deployment is anything other than becoming, also, more complex. How can you help have many diverse requirements, and a huge a macro site. Small-cell deployments include many mobile operators in dealing with the increase in number of devices. types of access nodes. complexity?

Densification is also about indoor DAS, it’s about Todd: At Ascom Network Testing, we have a outdoor DAS. It’s about heterogeneous networks, number of solutions that help carriers densify their

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |39| networks. One of them is in just identifying the That’s where we’ve always played. We’ve played need itself. in identifying the need, identifying the problem areas. But with the launch of TEMS iBuildNet We have TEMS Pocket, for example, which is earlier this year, we’ve now stepped more into the probably one of the most popular of our solutions game of not just testing to find out where the for testing, but we also have TEMS Pocket Remote. problems are and what the problems are, but now, “How do we design a solution that addresses With TEMS Pocket, I can go in and do a one-time those problems?” TEMS iBuildNet is that solution. test. I can either drive test, or do a site survey, indoors or outdoors. I can get an idea of how the Monica: That is interesting, because a small-cell network is performing at that moment in time, at deployment is more complex than a macro-only, that particular location, across those particular because you have the two layers: small-cell and location parameters that I’ve defined. macro-cell layers.

TEMS Pocket Remote comes into play here, From a testing point of view, it is a challenge, isn’t probably more than even TEMS Pocket by itself, in it? How do you deal with the increased that TEMS Pocket Remote is a stationary testing complexity? product. Basically, it’s an approach to testing the network over a period of time, at the same Todd: TEMS Pocket and TEMS Discovery both location. allow us to look across multiple RATs at the same time. It’s not anymore just about testing; it is now If I want to know, at any point in the city, or about planning and designing the solution for the multiple points in the city, or multiple points network, too. can do simultaneous multi-RAT design and indoors, or around the buildings, “OK, how is the simulations. Then, as a result, what you’ll have are network performing over a period of time?” TEMS That has to be planned and designed based on simulation measurements. Pocket Remote would give me an idea of the what carriers need. When they go out and do the capacity. It would also help me identify any not testing, they already have something in mind. What we have actually found, Monica, is that with spots and any hot spots. They’ve identified problems, or they’ve had the automated simulation capabilities of TEMS customers calling. Their network team has been iBuildNet, it has been proven to be within 5 dBm Then I would match that with TEMS Discovery. identifying specific areas. margin of error with actual network TEMS Discovery allows me to analyze that data measurements once the solution’s been deployed. and look at it geospatially, and begin to identify But it’s when they can say, “OK, these are the where those areas exist and what type of events capacity targets that we want to achieve in these That’s critical, as well. It is important to make sure are taking place to cause the issues that we might areas. These are the coverage holes that we have. that you’re not overdoing your design, not have identified with TEMS Pocket. These are the areas indoors and outdoors where overdoing your network. You don’t want to just we need to partner with the enterprises to make a throw an access node out there, or 10 or 15 of That’s the network testing piece. That’s where, at better quality of experience for our subscribers them, and say, “Well, boom! This is, this is, uh, Ascom, we come into play. It’s really part of and for their customers” – that’s really the going to work!” Or a small cell on a lamppost when identifying the need. planning design part of it. That’s where TEMS you may not need it on that lamppost. You may iBuildNet comes into play and you can do that. You

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |40| need it on a park bench down the road, or on top There’s always new things that are happening. and the virtualization of networks within of a building. What do you see coming up that is relevant to our themselves. operators, and I guess, to you as well, if you That’s where the planning and design comes in, choose to deal with it? I’m excited about that, but those are all things that and with TEMS iBuildNet, it’s not just indoor. TEMS we have to factor in as our engineers, as our R&D iBuildNet allows you to design hybrid indoor and Todd: LTE will continue to evolve, and obviously department, continue to develop new solutions or outdoor, outdoor to indoor, or just indoor or just what we’re looking at whether there is a advancements on solutions that we already have outdoor networks. competition between the carrier and enterprise in place – we have to take all of those things into Wi-Fi. The Mobile Broadband Alliance is certainly consideration. Monica: These are all tradeoffs that operators concerned about the carriers entering into that have to make. You don’t want to use a lamppost if 5 GHz space where Wi-Fi has played, and Monica: It’s a challenge, because, as you say, 5G is you don’t need it. Similarly, you want to understandably so. not defined yet, but you need to test to make sure understand whether it is indoor or outdoor it performs as expected before launch. capacity that you need. Some of the solutions will be hybrid scenarios, where carrier Wi-Fi and enterprise Wi-Fi play Todd: We want to encourage 5G, but we’re You offer a 360-degree approach, where you can nicely in the same sandbox. We have to certainly basically a supplier. Our main objective is to make plan, execute, and test. include that within not only how we test, but also sure that we’re partnering closely with, as I how we design a network. That’s one of the future mentioned, the carriers themselves. What are Todd: That’s it, exactly. Carriers don’t have money. things. their roadmaps? What’s going on with them? When I’ve been out at trade shows with carriers and talking with them, the one thing I hear over The other one is, of course, 5G is a big one. We’ve We’ve got to. I think that’s just good supplier and over from them, as well as from enterprises all heard about 5G. It’s on the horizon. relationships with your clients. It’s important to and commercial segments, is nobody has the Deployment is actually expected in that magic year make sure that you’re listening to them. And money. Everybody’s strapped by budget 2020. There are some already out there, saying, you’re not just listening to them, but you also have constraints. “Yeah, we’re deploying 5G, we’re testing it today,” people and teams, as we do, that are actually or playing around with it. dedicated to the various organizations and the With those budget constraints, who has time to various initiatives, including heterogeneous figure out, “Is it the right place -- is it the right time Until 5G settles down, we have to take a step back networks. for the solution? Do I need a small cell, or do I and say, “OK, well, we’re just going to stay in the need a DAS system? Do I need DAS, or do I need loop,” make sure that we’re connected to the Their responsibility is to keep a finger on the pulse Wi-Fi, or do I need both? Where should I place carriers, make sure we’re connected to those of what’s going on, what’s tested, where we are. them to make sure that my network is going to driving the technology to advancements in the 5G We attend all the events – we attend all the 5G perform to the target expectations that I planned arena, as well as the various associations and expos, all the HetNet expos, CTIA. As part of that, it to perform and meet those QoE expectations?” organizations that are driving all of that, like 3GPP. we’re not just going to exhibit our wares, but it’s also to network with everybody else that’s in the That’s what TEMS iBuildNet will do. But between the LTE-U, the advanced Wi-Fi on the industry, that’s driving a lot of these carrier side, and 5G in 2020, you’re also going to advancements. That’s what we are doing today. Monica: Going back to the technology that we have cloud RAN. That’s something we also have to have for densification. There are some new look at – it is the virtualization of network capacity Monica: You were mentioning 2020 before, for developments out there, so it’s a moving target. 5G. You think that there’s going to be one date

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |41| when 5G will become commercial? I’m getting the I agree completely with you. I think 5G is It’s a big conversation that’s going on now, it’s impression that it’s going to be a lot of incremental incremental; that’s why a lot of it’s going on. going to continue to go on, and I think, to your changes. There’re a lot of people already saying that they’re point, we’ll see it little bit by little bit. launching it. In fact, they may be, but is it really Todd: We all remember the 4G, when it came on going to be what’s finally called 5G? It’s very Let’s keep in mind, 2020 is just four years away. board. There were a lot of carriers that said, “Yeah, doubtful. Actually, less than four years. January 1, 2020, are our network is 4G,” but it really wasn’t 4G, it was we going to have 5G? Very doubtful that we’ll 3.5G. They called it 4G because it achieved 4G We’ve got several more conversations across the have the final version of 5G by that time. speeds. This is all about speed. Even 5G is all about various groups that define the advancements of speed, as well as efficiency. the different technologies, not just small cells and those that are driving those technologies, but also DAS and Wi-Fi.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |42| About Ascom Ascom Network Testing offers expertise and solutions that enable wireless operators to expand network capacity, improve operational efficiency and deliver a premier customer experience. Ascom develops its own line of mobile network testing, monitoring and post-processing solutions that have been trusted by mobile operators for decades under the TEMS™ brand. Today, those solutions enable field testing (drive, in-building, autonomous) of mobile networks, automated post processing of data collected via these – and other third- party – probes, OSS-based network troubleshooting and optimization, and application testing and monitoring. About Todd Cotts Todd Cotts is part of the Product Team at Ascom Network Testing. Todd has over 15 years’ experience in the telecommunications industry, having worked for some of the biggest names in the industry, including Sprint Nextel and Kyocera Communications. Having served in a number of strategic and leadership roles throughout the years, including directing product management of an indoor-outdoor network experience testing solution, and with an MBA in Marketing and near completion of a second Master’s degree (MS in Psychology with a Marketing focus), Todd brings to the table a multi-faceted approach to confronting many challenges faced by telecommunications professionals today.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |43| CommScope’s experience in DAS has focused both mobile operators face in moving the telecom Profile on indoor and outdoor installation. While the main infrastructure from macro sites to DAS or small focus for DAS is still on large venues such as cells – closer to the subscriber and in locations that CommScope stadiums, CommScope is expanding its solutions to are more visible, easier to access and difficult to make DAS more cost effective and easy to deploy protect. CommScope offers a wide range of network in smaller venues. infrastructure solutions that help operators and At the core of CommScope densification efforts is enterprises build, optimize and improve the CommScope offers the OneCell system – a hybrid the company’s extensive experience in optimizing efficiency of wireline and wireless networks. Its small-cell solution that combines features from antenna performance to deal with increasing solutions target: DAS, C-RAN and small cells to provide a scalable traffic density and the coverage requirements for solution for indoor environments with demanding voice and data services. CommScope invented Cellular and other wireless access networks capacity and coverage requirements that uses the remote electrical tilt (RET) solutions that allow Ethernet existing infrastructure for backhaul. operators to shape their antenna beams in real Wireless and wireline fronthaul, backhaul and time. It also offers a Six-Sector Turnkey Solution transport CommScope solutions include metro-cell that splits a single antenna beam into two, Cable networks concealment solutions for outdoor deployments. doubling the capacity, according to the company, These are designed to address the challenges that without the need to install a second antenna. Fiber optics networks Data centers IoT connectivity These solutions can be deployed both indoor and outdoor environments, ranging from stadiums and public transportation hubs, to medium-sized enterprise locations, to small-business and residential locations.

In the mobile space, CommScope provides wireless operators with tools to improve network and spectrum utilization by boosting speed, capacity and coverage in the RAN and in the backhaul. In this context, densification is a key focus that CommScope addresses primarily with indoor DAS and outdoor DAS, small-cell backhaul and fronthaul, and cell-site solutions. The recent acquisitions of Airvana and the Broadband Network Solutions (BNS) businesses of TE Connectivity have broadened CommScope’s portfolio in densification.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |44| The densification piece is key. That will be the In the building space and the oDAS outdoor space, CommScope number one way that you see this 1000X capacity we provide the DAS equipment, and the design come into the network over the next 10 years. At services, and sometimes even the deployment Building the CommScope, we have been doing densification for services to do that. operators for a long time. Densification is nothing new, but it will certainly accelerate. Monica: As you said, densification is nothing new. foundations for It’s accelerating, but it’s always been there. It In the past, we have started with the macro initially was more for coverage. Now it is more for densification towers, which are certainly the most efficient and capacity. Densification for capacity is different A conversation with economical way to start off for the operators. because the technology’s different, and because They’ve been sectorizing their sites, going from the the needs are different. Ben Cardwell, Senior Vice traditional three-sector sites to six-sector sites or President and Segment Leader, even more. We’ve been providing the tools and How do you see it changing, especially in the the know-how for them to do that. Once they get relationship between densifying the indoor versus CommScope Mobility Solutions to that point and they’ve sectorized their macro the outdoor? sites, then they look to more in-building sites. In- building has for a long time been a key avenue to Ben: It is changing. Operators will need to do both. Monica Paolini: Our conversation on densification densification. They’ll need to cover more and more buildings and RAN evolution at CommScope is with Ben from the inside out. In the past they have focused Cardwell, the Senior Vice President and Segment Second, CommScope has been the largest provider mostly on public-access venues – big, big venues Leader of CommScope Mobility Solutions. of DAS systems in the marketplace over the last where lots of people go. Those have been done. decade. We’ve deployed in countless public and Those are big buildings, big systems. Ben, can you tell us what, in this context, you do at private venues. CommScope? Going forward, we will need to move into those The third thing we’ve done is working with neutral second- and third-tier buildings that are more in Ben: As we all know, wireless networks will need hosts and operators to deploy what we call oDAS the 200,000 to 500,000 sq ft size. These new roughly 1,000 times today’s capacity in 10 years. networks, or outdoor DAS networks. oDAS can put systems will need to be much lower cost and be Operators generally have three tools they can use loads of capacity on street corners and other deployable by a different type of professional. to increase the capacity of their network. outdoor locations in a very small footprint. Maybe an IT professional instead of an RF engineer with 15 years of experience. We’ve been focused The number one is to add more spectrum. This is Monica: What is it that you personally do at on that in-building space, creating the platforms very expensive, and the opportunity to get more CommScope? and the ecosystem to allow DAS to scale to a much spectrum only comes along every several years. larger audience. Ben: In the sectorization of the macro towers, we Number two, they can make the spectrum they provide high-performance antennas, filtering Also in that space we’ve been developing have operate more efficiently. capabilities, and connectivity that allow operators world-class indoor small-cell solutions. We have a to sectorize their sites. When you do that, you solution called OneCell that provides a cloud-RAN Number three, they can densify the network, add require some very finely tuned cell boundaries, approach to in-building, giving maximum capacity more cells, and bring users as close as possible to a eliminating interference. That’s one thing we do utilization in a building. hot fiber hop-off point. best.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |45| In the outdoor space there will have to be many, many, many more of these small cells, especially in urban and semi-urban environments just to carry the traffic loads.

There are three major problems that the operators have in this space. It’s getting power to the site, getting backhaul to the site, and most important, site acquisition.

Site acquisition is becoming a major challenge. That’s something that, also, does not scale well in the industry, because it’s down to what the locality will allow. As we know, different regulatory or local approval councils have different requirements. We’re focused very much on creating solutions that allow scalability in this space.

Monica: You raised many topics that I want to follow up on. Let’s start with the outdoors.

You said that you’re trying to help operators districts in the US where you cannot see the Ben: I think there’s going to continue to be a large address power, site acquisitions, and deployment. equipment, you cannot see it at all. amount of customization, but the more of these How can you help them? cities and localities we deal with, through our We’ve done several historic districts where we’ve partners, we are learning some of the Ben: Site acquisition is, again, the number one replicated old-fashioned light poles that are from commonalities. We are working very hard to challenge. It’s all about concealment and not the early 20th century. There were gas lamps on create some solutions that will be a least common creating an eyesore for the public. the street, for example, to conceal the denominator, if you will, and more likely to be infrastructure inside and enable the fiber sited about various cities. We are creating cabinets and concealment connectivity to the site with our fiber solutions, solutions that enclose small cells, antennas and all and even backup power in the sites for fail-safe The other key to this is we’re seeing operators and the infrastructure required for a small cell. operation. neutral hosts forming public/private partnerships with the likes of energy companies, or other utility Small cells may be placed either inside a pole, Monica: Today, the requirements from cities are companies, or the municipalities that actually own under a park bench, in something that may look highly varied. Do you think they’re going to the poles. like a vending machine – something where it’s converge at some point, that we’re going to have a really disguised from the public. This becomes consistent set of requirements across You’ll see in many cities that the poles and the even more pronounced in the numerous historic municipalities? infrastructure tend to be common throughout that city or municipality. We can create, oftentimes,

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |46| two or three solutions that we can get approved Ben: They both have pros and cons. Really, it’s DAS solutions that we have are planned, installed, and deploy in those locations. about the specific task at that specific site. Again, commissioned, and operated just like any other operators will need multiple tools in their basket of piece of IT infrastructure. Monica: What is the relevance of the system tricks to do all the sites that are required. integrators in all this? We can enable an IT organization, an enterprise, to The advantage of oDAS is that you can bring loads use those. Those would still be an economical, Ben: At CommScope, we typically design and of capacity into a very small form factor. You can small, mini-DAS solution, if you will, where you manufacture the infrastructure to make this feed it with fiber. You can feed multiple operators. need multiple treatments of bands, or more than happen. The integrators play an absolutely key You can feed multiple frequency bands to that one one operator. role. The local integrators tend to know their radiating point, a very small form factor that can environment very well. Of course, there are some be hidden. This is the advantage of oDAS, if you In a building like corporate headquarters, where national operators in various countries, and don’t need that much capacity at a particular site. the enterprise may be standardized on a single integrators that understand the whole ecosystem. operator, all they need is one operator in the You only need a few channels or a single operator; building, and maybe one or two bands. Then a Generally, the local integrators play a key role then a small cell or a remote radio head, a small cell starts to make more sense. Those because they have worked in that environment miniaturized remote radio head, can be a very solutions, also, are becoming more enterprise before, either providing the solutions to the utility good solution. Really, it’s all about looking at the friendly because, to get in-building infrastructure companies or for the municipality. They’re very economics for the task at hand. to scale, you’ve got to tap into that enterprise important as part of the ecosystem. ecosystem. Monica: You want to have multiple tools to Monica: Do you think they’re going to play a more address the requirements. We have powerful DAS Monica: I guess there are in-between solutions, important role than operators themselves in doing for stadiums. If we go indoors, obviously that where you have a C-RAN approach. Do you the installations? solution is not relevant for a small office. consider DAS to be a type of C-RAN? Your OneCell approach seems to be a hybrid of small cell and Ben: In doing the installations, absolutely. Because How can you get DAS to adapt to the smaller DAS, with a C-RAN architecture. of the scale and the number of small cells or venues? What are the tradeoffs between a DAS, densification points, whatever type they may take even a DAS for a small office or venue, versus small Ben: The OneCell system uses what you might call on or that are required, operators are not going to cells? a hybrid approach. It’s not a stand-alone small cell. be able to do this all themselves. It’s a collection of access points within a building to Ben: We believe that DAS will continue to be the form a C-RAN. We call this a managed C-RAN They’re not going to have that scale with their own tool of choice for the large public access venues solution, an intelligent small-cell solution. resources. They’re going to have to rely on local and sporting stadiums, airports, the big high-rises, partnerships and even some regional partnerships big shopping malls, things of that nature, where With OneCell, you have a system that operates in a to get it done. every operator is going to want to be on that closed environment inside of a building, with the system. They’re going to have to provide every bit backhaul to the operator, that can be deployed Monica: You mentioned both oDAS and small cells of spectrum they’ve got into those venues. extremely cost effectively. for outdoors. What are the pros and cons of each solution? As we get to the smaller venues, we need new Monica: What kind of a backhaul do you use for tools. Again, in these smaller venues sometimes the in-building connectivity? we’ve been involved in DAS, obviously. The newer

Monica: Fiber from the building, and then is it Ethernet within the building?

Ben: Within the building it’s generally your standard twisted pair, Cat 5 or Cat 6 cable that is already in that building.

Monica: Operators are looking more and more into using unlicensed bands. They’re doing that already with Wi-Fi, but also they are looking at LTE unlicensed – LTE-U, LAA, MulteFire. What are you seeing or what are you doing on that front?

Ben: We believe that those unlicensed bands will be very important, because spectrum is one of the most valuable assets. Using that unlicensed band will be key. We believe that this signaling, the part of the network that controls the call, will always be in the licensed band, because the carrier needs to control that very carefully.

The newer technologies, like LAA or LWA, will allow the LTE protocol to take advantage of some of that unlicensed spectrum, grab it, use it while it’s available, and be able to ramp up and down the use of that dynamically. creating DAS solutions in antennas and on capacity, for that matter, in an unlicensed band is We think that’s going to be very key to have that connectivity infrastructure to minimize all the an entirely new endeavor. kind of on-demand use of unlicensed spectrum. interference in the building. When you do that, it adds to the complexity of the Ben: You’re exactly right. RF network. Our expertise at CommScope is In the outdoor space, the same thing. We’d be managing that RF interference. working to create the antennas and filtering Monica: Many operators are a little bit nervous solutions to make that sufficient. about how to do it, because still they have their We believe we’re the best there is at licensed band as an anchor, but there are still a lot understanding and mitigating the interference in Monica: That’s interesting, because interference is of challenges there. What are you hearing from the RF layer of the network to make all those nothing new, but for a mobile operator to manage operators? bands work together efficiently. We would be interference in an unlicensed band, or to manage

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |48| Obviously, interference in an unlicensed band is Are they going to pay for infrastructures Monica: Yes, absolutely. Scalability is crucial here. something you can manage only to some extent. It themselves? How is that going to work out? One other trend is virtualization. Is virtualization depends on what else is going on at the same going to help scalability? time. Ben: This is a very good question. That is extremely important, that we enable that enterprise Ben: It will help scalability. It will definitely help Ben: That’s correct. Interference has become a ecosystem to be somewhat self-sufficient. improve the ability to deliver 5G services. It’s a much more serious and pointed issue now that journey to virtualization for the operators. They’re LTE has been deployed. In spread-spectrum The carrier is going to always have to play a key starting with some very simple things, like systems like LTE, and even in CDMA and in UMTS, role, because it’s the carrier’s licensed spectrum centralizing the baseband in the network. as we move up in the G cycle – 2G, 3G, 4G – we’re talking about. They will always have to be interference becomes more and more important. comfortable with whatever is radiating on their Today, in most cell sites the baseband is sitting at spectrum. At the end of the day, it’s their the bottom of every tower, dedicated to that One reason is that the traffic on the network is customers. If my cell phone doesn’t work, I’m tower. Sometimes it’s heavily utilized. Sometimes getting higher. The number two reason is that going to call my operator and be the one that it’s not utilized. If you take a cluster of 50 sites, and you’re trying to get more and more speed out of complains. We’re working very hard to create the take that baseband away, move it to a centralized this fixed-frequency resource. You’ve got to fine products that are friendly to the enterprise but point, you can make it available to all those 50 cell tune and clean up that spectrum as best as we acceptable to the operator. sites dynamically. can. Every antenna’s got to have a very clean, targeted pattern covering the areas, not More importantly, we create products that are When there’s a football game going over here, you overlapping with its neighbor. extremely simple to design, deploy, and install, so can shift all the traffic over there. If it’s rush hour that the much larger market of IT engineers who and a traffic jam, you can shift it all over here. We We’re managing that type of interference plus just are going to be installing them are comfortable see that starting to happen already today. That’s the normal interference that can come from the working with them and can install them reliably. step number one. environment of operating a mobile network. We’re creating products that are easy to install, It’s not just about the IT people in the building. It’s The next step would be, from a capacity that work well together so that – even in the about the ecosystem that feeds them today. We perspective, virtualizing that. Moving it further interface points between the different pieces of have to enable that ecosystem because, again, back in the network and putting it on software- equipment in the network – you are not creating we’ve got to scale to not just doing the thousands based servers. That can even be made available opportunities for an installer to create interference of buildings a year that are done today with DAS. dynamically to larger chunks of the network. when he connects them together. We’ve got to scale this to tens of thousands or even to hundreds of thousands a year of buildings We think this is a migration that will happen over Monica: As you say, one other crucial thing, as you that can be deployed. the next five years as people prepare for a barrage move indoors especially, is that you need to be of new traffic. friendly to the enterprise. They need to be able to We believe there’s five billion sq ft of opportunity easily fix problems if there is something wrong, or out there, of uncovered commercial space out Monica: What are you working on at CommScope to have an integrator come in. there that needs to be covered. The only way right now to meet the challenges over the, say, we’re going to do that is by enabling an next five years? How do you see the role of the enterprise in all this ecosystem. We’re working very hard at that. changing? Are they going to be more involved? Ben: At CommScope we have two key technologies that are going to be the

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |49| fundamentals of the 5G network. In the next We’re creating solutions that are going to drive Things like massive densification will become a big generation, network broadband and wireless this centralized C-RAN and virtualized RAN (or play where you have beamforming, because networks will converge, and customers won’t be vRAN) network. There’s going to be a lot of fiber in beamforming helps further reduce that able to tell, and won’t care, whether they are on a these networks. Very few people know how to interference in the network. fixed network or a wireless network. They’ll get manage fiber and wireless as well as we do, as it the same type of service from both. relates to a wireless network. On the in-building front, designing the next- generation DAS systems and small-cell solutions We believe that wireless and fiber are going to be Then, at the very network edge, at that wireless that scale to that enterprise ecosystem we’re the two big enablers of that converged network. access layer, we are working with our industry talking about – we think that’s a big opportunity We are the leaders in both in the layer of the partners to create the same types of infrastructure that’s largely untapped out there. network that we operate. we have in the past. Smarter antennas, where they have more intelligence in them.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |50| About CommScope CommScope helps companies around the world design, build and manage their wired and wireless networks. Its vast portfolio of network infrastructure includes some of the world’s most robust and innovative wireless and fiber optic solutions. Its talented and experienced global team is driven to help customers increase bandwidth; maximize existing capacity; improve network performance and availability; increase energy efficiency; and simplify technology migration. You will find CommScope solutions in the largest buildings, venues and outdoor spaces; in data centers and buildings of all shapes, sizes and complexity; at wireless cell sites; in telecom central offices and cable headends; in FTTx deployments; and in airports, trains, and tunnels. Vital networks around the world run on CommScope solutions.

About Ben Cardwell Ben Cardwell is Senior Vice President of Mobility Solutions at CommScope. He is responsible for leading the global business unit that develops innovative wireless solutions for use in service provider and business enterprise networks around the world. Prior to his current role, Ben was senior Vice President of Global Wireless sales, responsible for leading all of CommScope’s customer-facing activities for the wireless group globally. A 25-year veteran of the telecommunications industry, Mr. Cardwell also served as Vice President of Wireless Sales for CommScope Asia Pacific, where he worked closely with US and Asia Pacific service providers in developing and deploying radio frequency and data communications infrastructure for voice and data services over mobile networks. Prior to joining CommScope, Mr. Cardwell served in various leadership positions in research and development, product management, systems engineering, and field sales with UTStarcom, Ericsson, and 3Com. Mr. Cardwell graduated from Davidson College in North Carolina with a bachelor of science degree in physics. He holds an MBA from Kellogg Graduate School of Management, Northwestern University.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |51| Ultra Mobile Broadband, a 5G access other sports events, political conventions, or Profile platform being developed to bring high other public gatherings. capacity and low latency to mobile networks InterDigital that use frequencies above 6 GHz. Ultra Next Generation Networks Platform, a Mobile Broadband leverages new spectrum flexible routing solution to increase efficiency Since 1972, InterDigital has developed a wide array allocations in the millimeter-wave bands, and in transmission and optimize video of new technologies and contributed to it can target outdoor hotspots and indoor performance, leveraging NFV/SDN, MEC and establishing wireless standards across cellular and environments, as well as provide in-band ICN. The platform is a hybrid approach that IEEE 802-based technologies, with a strong focus backhaul. It supports both MIMO and multi- combines features of IP and ICN. It uses on Wi-Fi. user MIMO (through hybrid beamforming) in multicasting to increase network utilization TDD spectrum. Because of the short range of and resiliency and to reduce latency. In a Complementary to its work on wireless, millimeter-wave technologies, deployments move away from host-to-host centralized InterDigital has developed IoT solutions for the will target high-density locations, such as bus processing, this approach pushes content and transportation, automation (e.g., sensors, remote stops, central-city streets, parks, malls or functionality toward the edge, closer to control), healthcare and industrial verticals. In this airports, or specific events, such as marathons subscribers. area, InterDigital has launched wot.io™, an and operating environment that facilitates the development and launch of IoT applications in which multiple ecosystem players, platforms and interfaces interact. The oneMPOWER platform also supports growth in IoT by providing tools to manage devices across verticals, applications and form factors.

In wireless, InterDigital focuses on three solutions, all supporting the densification trend among mobile operators:

EdgeHaul™, a low-cost, high-capacity, millimeter-wave transport system in the 60 GHz band, which can provide mesh backhaul in small-cell and carrier Wi-Fi deployments. EdgeHaul can also be used for residential fixed-wireless broadband access. Low latency, support for network slicing, and SDN support make EdgeHaul a platform well suited to 5G deployments. The low latency and high capacity make EdgeHaul a solution that can be used for fronthaul in C-RAN deployments.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |52| Alpaslan: In the 5G domain, we distinguish spectrum bands, where you have much wider InterDigital between bands below-6 GHz and above-6 GHz. reach. But it’s also an advantage, because it The primary focus in my team is on bands above-6 reduces interference. It’s a tradeoff there, because Densification in GHz. Above 6 GHz deployments are in the you can reuse spectrum more regularly than you millimeter-wave bands. Some people refer to can, for instance, in the 700 MHz bands. bands above-6 GHz as centimeter-wave and millimeter-wave millimeter-wave bands. We are talking about Alpaslan: Yeah, definitely. When you look into frequencies in the 28 GHz or 39 GHz range. We’re millimeter-wave bands, you need highly bands talking about spectrum in the 60 GHz or 70 GHz as directional antenna systems. With the highly A conversation with well. directional antenna systems, people think that you definitely suppress the interference. However, Alpaslan Demir, Principal The beauty of these frequencies is that the new that’s not very accurate. Overall, you can reduce Engineer, InterDigital bandwidth they make available is tremendously the interference. You still have interference issues large. You are talking about Mbps or multiples of- with highly directional beams. 100 Mbps bandwidths, with up to 2 GHz Monica Paolini: Our conversation with InterDigital bandwidths, or multiples of 2 GHz, especially at Monica: What kind of deployment scenarios do about densification and the evolution of the RAN is 70 GHz. This gives you the ability to increase you see for these millimeter waves? with Alpaslan Demir, a Principal Engineer there. densification. The more bandwidth you have, the more data you can transmit. Alpaslan: You definitely can talk about indoor and Alpaslan, could you give us an introduction on even outdoor scenarios, urban environments, what InterDigital is doing on densification and When it comes to the millimeter-wave domain, definitely – wherever you need very high density, what you’ve been working on, on that topic? one thing we need to understand is that, due to you can utilize the millimeter-wave network. the propagation loss, you need to deploy nodes Campuses, enterprises, malls, all sorts of locations Alpaslan Demir: InterDigital is a multifront very closely. Maybe you’re talking about spacing can definitely utilize millimeter-wave applications. company for many research topics. One of the nodes at 50 m to 100 m to 200 m. This topics we are heavily involved with is densification. automatically creates densification. Monica: You mentioned 5G. We’re seeing a trend We support multiple projects in the densification towards some of the elements of 5G becoming domain. When you look at the bandwidth in terms of Mbps available before true 5G is deployed. Do you think per hertz, the concept of areal densification comes this is the case? If so, what is the timeline? Do we I’m part of the team who is looking at the next- in. With the addition of millimeter-wave bands in need to wait for 5G to be completed, or can we generation millimeter-wave network. We look into densification, the definition of capacity should not get started ahead of 5G full definition, with gradual the challenging problems in densification and work be limited to bps/Hz but it should involve space as changes within 4G? on creating solutions. When we talk about another dimension. For example, if there are 50 densification, we talk about ultra-dense networks links deployed over one sq km, each with 10 Gbps Alpaslan: That’s a very good question. In a 5G in general. over 2 GHz channel bandwidth, then the total domain, certain technologies are developed and capacity can be defined as 500 Gbps/sq km. certain prototypes are available. However, when Monica: There are different ways to go about you look into the ecosystem, you do not see densification. Operators are pursuing many Monica: The need to have a denser network is a commercially viable formats yet. Especially in the different roads to get to a massively densified challenge in millimeter-wave bands. It’s a RF domain, the antenna structures, solutions to network. One way is to look at new bands. limitation compared to the below-5 GHz cellular

This is delaying things. The other aspect is the standardization process. As they did in the 3G and the 4G domains before, now companies have started looking into the standardization aspects of 5G. A reasonable time everybody is talking about for commercially viable deployments in the 5G domain is 2020.

Monica: Before that, before we get to 2020, what do you think is going to happen in terms of densification?

Alpaslan: There will be prototypes, there will be a ton of research, and the testing and the capabilities are going to come up with densification. In Europe, there are several research facilities or places where the deployment scenarios are being tested.

In the United States, we are expecting similar activity. The FCC is making new spectrum bands As engineers, we are definitely up for the But 5G is not just a single concept that fits available in which we can start testing and challenges, and we’ll find solutions to everything. 5G has many facets. We’re talking deploying our prototypes. accommodate whatever the needs are in this about low-latency applications, such as those for domain. tactile internet. We are talking about high- Monica: In terms of regulation and band throughput applications. And there are many availability, do you expect to have millimeter-wave Monica: What about the standardization process additional aspects. Even IoT concepts are panning bands that are available worldwide or in most you mentioned? 3GPP is working on this, and out as part of 5G. The 5G is much, much greater markets? there is a lot of research going on. Do you expect it than 4G and 3G. But we definitely need to have to be polarized, or is the industry converging nicely consensus. Alpaslan: It’s very difficult to achieve a global on new standards? synchronization on these bands, and everybody Monica: IoT is a driver to 5G and also benefits knows that. For instance, the 28 GHz band will Alpaslan: There are always some glitches here and from it. How important to do you think IoT is as a become open in the United States, but you may there. However, when you look into how 3G driver for densification? not have the same band all over the world. That’s panned out, and 4G similarly, I personally don’t a challenge. see much of a problem converging on 5G-domain Alpaslan: My answer may be a little bit technical solutions. here. In normal densification, the concept of fiber

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |54| drops for a dense network may become an issue. With respect to the number of fiber drops in the system, how could you reduce this number of fiber drops?

Let’s say that instead of using fiber drops in every node, you can connect these nodes with wireless backhaul. You can obviously increase the number of hops doing that, but the main benefit is the total cost of the deployment. The backhaul network that you create will be wireless. However, due to the multiple hops that you introduce, there are latency problems that you have to solve.

From any standardization perspective, the standard has to be cognizant of all these scenarios and find a common ground, so that it empowers people to use their own applications and interpretations, and it creates harmony.

Monica: Because there is so much spectrum in the millimeter-wave bands, can you use the same band for both backhaul and access?

Alpaslan: Yes. One of the approaches we take at InterDigital is that we’ll have these two GHz links available to you. Not always do you need the backhaul links utilizing these bands. In our approach, we’ll definitely look into the joint access Monica: Can you tell us something about how you performance, the radio performance, and the and backhaul concepts. Where we deal with manage interference? leakage that may come from other channels. interference management, you do have interference from the backhaul to the access or Alpaslan: One simple example I can highlight is Another example is that interference metrics from access to the backhaul. that, if you have multiple channels available to you knowledge of the directionality of interfering links between neighboring nodes, you can definitely set allow you to create a dynamic routing setup to On top of this, you have latency issues that you these nodes using a different channel, such that mitigate interference. You can even create have to fix. We look at densification holistically, the neighbors will not see each other’s signals and different directions. Instead of going through a hop and enable the joint-access backhaul, and solve the impact of interference from the second-tier directly, you can use multiple hops to convey your many problems coming with it. neighbors will be lower. When you are doing this, information, so that you don’t interfere. This is you have to be very careful about your system’s basically with one link active. The one link is no

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |55| longer affecting the other link if you start playing What’s your view on what’s the best way to go takes longer to converge and implement new with the directionality of the links. about it? standards.

Monica: If you have multiple hops, latency may Alpaslan: I definitely see legacy devices as an InterDigital has resources working in both become an issue, and latency is a crucial factor in integral part of moving forward. We cannot just directions, cellular and Wi-Fi. defining performance. You need not only capacity, forget about legacy devices and the legacy but low latency to get a good quality of infrastructure, and then start building completely Monica: When I talk to them about higher experience. brand new 5G infrastructure. frequencies, they’re still a little bit nervous. They’re much more familiar with the lower In 5G, that’s going to be a major requirement. How We need to develop technologies that take frequencies, where they have full control and have can you achieve these lower latencies? advantage of existing legacy infrastructure, as well fully licensed bands with good penetration and as of the opportunities created by the bandwidth, wide coverage. As you move to higher frequencies, Alpaslan: We do have many other ways of solving the latency, and such of the 5G domain. the regulation is different, coverage is different. this problem. One simple example, again, is with What are you hearing from them? multi-hub systems. In the 5G domain, the systems are heavily populated, and highly directional beams are Alpaslan: Let me state this: there is no escape You are receiving packets. Normally, what you do utilized. But you cannot have coverage in every from creating solutions for millimeter-wave or is you absorb all the packets, and once you have corner. Even your hand movements or gestures on high-frequency bands. The overall industry is received everything in that packet, you start your handset will completely destroy your link, and pushing for that. We definitely see some operators forwarding. If the packet is causing you 10 units of you may lose connectivity. taking advantage of that, as well. Even today, we delay, now you have to wait for 10 units, and then talk about operators embracing Wi-Fi and start forwarding that same packet of 10 units to What are you going to do if that’s the case? Your connecting with Wi-Fi, and we do see a lot of ways the next tower. system has to be adaptable enough to go back to to merge LTE and Wi-Fi systems. the legacy network and start using it. Wherever it What you can do is create much smaller units makes sense, we should look for solutions that Moving forward, the anticipation is that some of within that packet, and start sending these small aggregate both the 5G and the legacy systems. the bands on the millimeter wave could be units without waiting for the entire 10 units licensed, or could be semi-licensed, and some of coming in. Let’s say you start transmitting every Monica: 5G, 4G, 3G, they need to be integrated the bands, like 60 GHz US ISM band, could be unit without waiting. You immediately reduce your tightly, so that you can have devices working back unlicensed. latency by doing this. and forth among interfaces. Wi-Fi’s part of the picture, too, right? From an operator perspective, it is best if you can Monica: In addition, with the increased complexity take advantage of what is out there. The operators in the RAN, there is much more work that we all Alpaslan: Wi-Fi is always part of the picture. As a should be more adamant about creating solutions. need to do in order to make sure the performance matter of fact, Wi-Fi standards are adopted much Some of the applications in 4G were unforeseen. is high. faster than 3GPP standards, because the focus is When 4G was enabled, then these applications more narrow and the scope of standardization popped up. However, inevitably, as you move to 5G, you’re more limited. When you look at the 5G domain moving to an environment where there are within the 3GPP perspective and, more generally, With 5G, what we have is a tremendously large multiple wireless interfaces used for coverage, at the evolution of the cellular domain, the bandwidth, a very fast network, and all these capacity, or both. You need to integrate them all. industry is much larger than the Wi-Fi industry. It

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |56| capabilities. A ton of new applications that would Alpaslan: For the densification purposes, we will introducing products, as well. We will do this not be mind- boggling would come up. We will see. continue to look into the solution space and only for 5G, but for what comes after 5G. solving the challenging problems. I see InterDigital Monica: It will be interesting to see. There’s a lot providing solutions, helping to shape the industry, Monica: 6G? of work ahead. If you look at the next five years, contributing to the standardization process, and what is InterDigital focusing mostly on? Alpaslan: 6G, hopefully.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |57| About InterDigital InterDigital, Inc. designs and develops advanced technologies that enable and enhance mobile communications and capabilities. Since its founding in 1972, its engineers have designed and developed a wide range of innovations that are used in digital cellular and wireless products and networks, including 2G, 3G, 4G and IEEE 802-related products and networks. For over four decades, InterDigital has been a pioneer in mobile technology and a key contributor to global wireless standards. InterDigital’s team of more than 170 engineers – approximately 80% of whom hold advanced degrees, including over 50 PhDs – has unparalleled expertise in major mobile connectivity and content delivery technologies. Since 2000, InterDigital has spent over $1 billion on technology research and development. InterDigital is a registered trademark of InterDigital, Inc. EdgeHaul, oneMPOWER and wot.io are trademarks of InterDigital, Inc. About Alpaslan Demir Alpaslan Demir is a Principal Engineer in the Future Wireless Group at InterDigital. He is part of a team working on NextGen Millimeter-Wave Architectures and currently focusing on activities related to millimeter-wave densification. He has been serving the wireless communications industry for more than twenty years with a unique combination of experiences relevant to MAC, PHY, and RF design. Notably, he is a prolific inventor with 56 granted and numerous pending patent applications to date. He is an IEEE member and holds a Ph.D. degree in Communications from Polytechnic Institute of NYU.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |58| a manhole cover–type lid. By placing antennas and macro-antenna concealment solutions for urban Profile BTSs below street level, operators can deploy and rural environments, Kathrein provides ways to unseen infrastructure that meets municipalities’ embed antennas into a variety of street-furniture Kathrein requirements more easily, and can have easier elements. Kathrein Inside Connect, for instance, access to backhaul and power, than other outdoor was developed in collaboration with JCDecaux to Founded in 1919 and headquartered in Germany, equipment. provide a modular antenna that can be integrated Kathrein is a leading vendor of antennas and into advertisement infrastructure (e.g., billboards related equipment, such as filters, combiners and Kathrein offers antennas for above-ground small and bus stop shelters). concealment solutions, for wireless networks and cells, as well. These are designed to operate in the for multiple verticals. increasingly complex networks that result from the For indoor environments, Kathrein offers DAS and growing adoption of small cells and need to small-cell solutions for different types of building Transportation is a vertical in which Kathrein has a accommodate higher and more diverse traffic structures and traffic patterns. In this area, strong presence, with antennas and other loads. And Kathrein continues to optimize them to Kathrein has recently introduced K-BOW, a micro components for the railway and automotive meet increasingly stringent requirements in terms C-RAN system, in which antennas are connected to industries. The company is leveraging its expertise of, for example, regulation power levels, a shared baseband, typically located within the in this area to enter the IoT market and, more provisioning, operations, antenna visibility, RF building. K-BOW supports multiple mobile access specifically, to provide connected car solutions. exposure, and ease of installation. frequency bands from 700 MHz to 2.7 GHz, Industrial automation, logistics and retail are other including LTE-A and Wi-Fi. IoT core verticals to Kathrein. One area that is growing in relevance to densification is antenna concealment. Alongside Kathrein also provides antennas for indoor connectivity of private networks in enterprise, hospitality and other venues. Those solutions can be used for content delivery (especially video) and other data services, as well as in satellite communications and broadcast.

Cellular networks, however, are the largest segment at Kathrein. The company offers a wide range of solutions for indoor and outdoor coverage, targeting different network layers – from macro cells to small cells and to DAS. To support operators’, need to boost coverage and capacity, Kathrein has developed multiple solutions that enable densification.

The most recent is Kathrein Street Connect™, a small-cell antenna system in which antennas are located just below street level, accessible through

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |59| We’ve made major strides that way, and relocated “where can we install antennas for small cells in Kathrein our office to the heart of the Telecom Corridor™ in new and novel places?” Richardson, Texas. We recently closed a Making densification manufacturing plant in Oregon, opened a huge They actually came up with the concept of new facility in Mexico, centralized our installing it in the ground, in what looked like warehousing in Texas, and become more involved manhole covers. They looked at a few antenna equipment go with industry working groups such as 5G Americas vendors and ultimately chose Kathrein to bring the and IWPC. solution to market due to the high level of quality unnoticed and engineering expertise. We did some joint A conversation with Jim Monica: But the company is based in Germany, development, and 17 trials in and around with clients worldwide. Switzerland to determine performance and DeKoekkoek, Product Line coverage capabilities. Manager, Antennas and Small Jim: Exactly. One of our challenges, in the past, was that we were more Eurocentric than we really Voila, we have this amazing product now that Cells, Kathrein wanted to be. We recognize that the US is a strong provides a solution zoning and aesthetic market for us and have taken some big steps to challenges all mobile operators face. realize the strong opportunities here. The Kathrein Monica Paolini: This conversation on the name means a lot to people. The quality is Monica: You said that they look like manholes, but densification of RAN and its evolution is with Jim excellent. It’s an enjoyable company to work for, they’re not. What’s the difference? DeKoekkoek, Product Line Manager for Antennas for those reasons. and Small Cells at Kathrein. Jim: The average person walking by can’t tell the Monica: When we talk about densification, there difference. The cover, of course, is not metal, like a Jim, could you tell us what you personally do at are many ways to go about it. Operators will, of normal manhole cover as that would cause Kathrein, and what Kathrein is doing in the area of course, pursue multiple paths at the same time. interference, but it is a metal-looking composite densification? But these days when I think about Kathrein, the cover. It’s designed and tested to withstand up to first thing that comes to mind is Street Connect™. 40 tons so it can handle heavy trucks and traffic. Jim: My role here is one of being the interface This is a novel approach to densification. between our engineering department and the Monica: What if you were to put them in a market. As part of that, it really is a good chance Jim: Street Connect has been a lot of fun for us. manhole itself? Obviously, you’d probably have for me to visit with customers, find out what their We were contacted originally by an operator in propagation issues with the metal of the cover, but needs are, where they’re going, what the Switzerland, by the name of Swisscom. They had does it have to be in a different area? directions are – and coordinate that with our done quite a bit of evaluation of costs related to engineering folks, who are largely based in deploying small cells. Jim: It does need to be in its own separate Rosenheim, Germany. enclosure. We don’t want to restrict access to an They had the typical problem where the data actual cover, where there actually might be a need At Kathrein USA we had big changes in the last usage is going sky high. They said that, since 2008, for workers to go into that. Also, it’s smaller in couple of years. We recognize that in a it had gone up a hundred times. They have very diameter. It’s about 17½ inches in diameter, which competitive business, Kathrein really needs to strict zoning standards in Switzerland and actually is too small for most people to get into. become a player in the US more than it has been. therefore difficult to deploy. They were looking at

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |60| Monica: Usually what you do when you go from Monica: If fiber is available, it’s easier to have to macro to small cells, you go from way up high, and the small cell underground than if you go to a you get progressively closer to the subscriber. lamppost, where fiber connectivity is usually a Kathrein goes one step further, because Street problem. Connect goes below the subscriber. Jim: Exactly. What does that do in terms of the interference, say, with the macro, when you have a co-channel Monica: Swisscom must have been talking to cities deployment where the macro and the small cell and municipalities. Were they open to this type of use the same spectrum? deployment? Streets are under the control of municipalities, and so you need their support? Jim: Placement is really key there, to try to prevent that interference. If you’re designing a system, Jim: Absolutely. The right-of-way for locations for you’re going to look at maybe trying to shield from the Street Connect is a big deal. Zoning and the macro cell using surrounding buildings. Part of placement of antennas, particularly in big cities – the coverage of this system really depends on where you have this huge, dense traffic load reflection. It’s commonly located adjacent to requirement – is a big problem. buildings, where we actually get good multipath, and good coverage that way. One of the fun parts of the Street Connect is that we have cities that actually are interested in Monica: In terms of cost, how does Street Connect deploying it and that have control of the right-of- compare to a lamppost or other outdoor small-cell way. We have customers who will advocate for it deployment? and streamline that zoning process.

Jim: Swisscom compared Street Connect to, first, Monica: Small cells represent a new approach to macro cells, and then to conventional small cells wireless deployments. We are trying to put the on lampposts and so forth, over a five-year period, telecoms infrastructure in a place where, to calculate all their maintenance costs, and the traditionally, there was none. installation of fiber. Swisscom found that it comes out to be about half. It was very attractive from a Are there other places? You’re not just deploying perspective of cost. small cells underground. Where else will your products go in the future? Really, that is the main driver here. Certainly, they’re going to use other tools as well, where it Jim: There’re a lot of tools in our toolbox. That’s an makes sense. I’m sure they’re going to still old cliché, one of our most popular to date for continue, as any operator would, with lampposts, neutral host and mobile operators is our canister and other types of small cells. But this really small cell product line. That’s one area, and we’re worked well from a cost perspective. constantly innovating and enhancing those. by adding features and capability – extended frequency ranges as new frequencies, such as the

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |61| 3.5 GHz and the 5.8 GHz, start to come into play. A between telephone poles. We have a kit to mount Kathrein is a huge supplier of automotive antennas lot of demand for that. the antenna hanging from the wire and next to a as well, so it’s a natural tie in. As the supplier to radio. We can accommodate 2x2 or 4x4 MIMO most of the major manufacturers in the world The other thing is that most of the small cells have with it. we’re already thinking about how we’re going to been 2x2 MIMO. Traditionally, that’s been support connected car, leveraging those antennas. adequate for most capacity requirements. But Again, we can adjust where the beams go. Those we’re starting to see that in major cities, where are the kinds of things we’re doing with the Of course, with the IoT, there are many layers. they don’t have any more capacity or place to put existing antennas. In terms of new products, we Kathrein also has a broadcast solution – in fact we another antenna that 4x4 MIMO is becoming a are coming out with a series of antennas that I have been providing these since the 1950’s. When reality, to get the capacity they need. We’re would liken to Lego blocks. They’re antennas that you have slow data applications, we may tie them adding the 4x4 MIMO. are designed to be mounted in hidden places. into broadcast facilities. We have satellite solutions Places like behind, say, an advertising sign in an and hybrid fiber coax. All of these tie together to Another approach in terms of antennas, both new airport or some public space. create a connected solution. We’re trying to and existing, is working with third-party firms to coordinate our efforts in all of those things for IoT. come up with flexible mounting and concealment You can configure them by combining them. If an options. operator needs more gain, or more directionality, Monica: How do you see broadcast fitting into the a combination of two, or three, or four of them IoT expanded usage, congestion, small cells? We have an antenna that we call a smart-pipe can be easily be placed together, or back to back. antenna, internally we smilingly call it the barber- Lots of good applications, so we’re really having Jim: Great question. Broadcast spectrum will be pole antenna, because that’s what it looks like. fun with the small cells. auctioned via the FCC repack. Remaining That antenna can be mounted on, say, the corner broadcasters after the auction are evaluating the of a building. Monica: Operators require a lot of flexibility. ATSC 3.0 standard that will allow 2-way They’ve got more bands, but also they have more communication in their spectrum. If implemented What’s cool about it is that the antennas can be services to support. and adopted, this will provide an enormous rotated, and this way you can customize the amount of bandwidth for IOT applications and direction of coverage. An operator can have this One class of services that is going to create a wide backhaul for wireless carriers. antenna on the corner of a building, in a range of demand requirements and new traffic downtown canyon environment, and aim one loads is IoT. There you have different types of Broadcasters will more than likely follow what the antenna to the north, and the other antenna to services, different type of devices and use cases wireless carriers are already doing – densifying the west. driving your opportunities. How is IoT going to their networks with small cells. impact what you do, and how can you help Another thing I can do with that antenna option is operators get into that market? Monica: Broadcast, it’s a concept we have been to mount it like a paper towel roll, meaning that working on for a long time, in terms of how we it’s horizontal. Mounted on the side of a building, Jim: Good question. Boy, IoT is coming fast. It’s actually do it. There is a lot of potential, but what’s an operator can aim one antenna at the lower coming faster than we maybe thought it was going the best way to harness it? With the increase in floors, and one antenna at the higher floors. to. Things like the connected car, for example. demand, we’ll have to see how that plays out.

Another creative solution that we’ve had fun with We are starting to build the 5.9 GHz, for example, Jim: It also could be things like what they call LoRA, is to mount the smart-pipe on what they call the into antennas so we can participate in that which is the 900 MHz. We’re also seeing L band, messenger wire -- the strand cables that run connected car business when it occurs. the 1500 MHz. All of those frequency blocks start

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |62| tying into the internet of things. You start using Really, placement is key. Adjustable them in the most appropriate way for whatever beams are key. Probably adjustable layer you’re trying to accommodate. power on radios is key. You can optimize a network that way to Monica: IoT would require additional bands. Do minimize those overlaps. Every you think there are going to be separate bands situation is going to be custom reserved to IoT, and specific applications with IoT – tuned, so to speak, as you try to say, connected cars, or something else? optimize the network.

Jim: They all have appropriate usages. From meter Monica: You mentioned at the reading to applications with low-latency beginning that you have been requirements, you’re going to utilize more than focusing personally on the US, and one technology. Where you’re reading a meter listening to what you hear from once a month, you don’t care when it comes in. your customers. You’re also When you’re trying to maybe do an emergency working for a company that’s based broadcast, you might have a different technology. in Europe. Can you tell us a little They all are going to be useful tools down the about what it is you hear that’s road. different in these geographies?

Monica: The availability of new bands enables Jim: In different places in the world, really, the We’ve done a lot of testing here in the US and operators to have small cells and micro cells main difference is spectrum availability. That’s one have active trials going on. We’re very working on different channels. This would simplify obvious area. There are different regulations in conservative in our RF exposure, the maximum life for everybody involved, because you do not different countries. In Europe, for example, the permissible exposure. Yet we’re trying to make the need to coordinate small and macro cells. height of an antenna is much more tightly system work well. We’re trying to get as much controlled. input power as we can, because we want to get At the same time, we still have a lot of co-channel good coverage and good performance out of it. deployments, where macro and small cells are In Europe, they also will tend to limit what they call using the same spectrum, and interference has to the lower-side lobe. They’re more concerned than Those are all the things that we’ve been doing in be managed. That has been, for a long time, a we are in the US about RF exposure to people. our trials. We’re really figuring out how to prevent major concern for operators. How can we address They tend to have really stringent standards for RF exposure, or interference with the macro that? that. antennas.

Jim: It is a tough question. The more frequencies In fact, that was an interesting part of the Street We’re ready to move forward, actually. We’re at a you put into usage, the more multiband antennas Connect Solutions. Switzerland has one, if not the point now where we’ve made good progress we have, the more combinations for most, stringent requirements for RF exposure in there. We’re ready to actually start installing intermodulation that you have. That’s one aspect. the world, by about 10 times. That was actually a these. Then again, you’re trying to deal with not great proving ground for the product in terms of interfering with the macro level. that RF exposure issue, which we really had to deal Monica: That’s exciting. One other thing is that with. with small cells, it seems like the technology is there – we’re all ready to go. All the vendors are

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |63| ready. Most of the time, the big challenge to the permissions, the zoning, and so on. Again, it Monica: This is all new. With small cells, we have rollout is really the operational part – installing it, seems like many of the larger cities that need this the need to have telecom infrastructure getting all the permissions right away, electricity, coverage are being very cooperative with us. We everywhere. But you want it to be less intrusive power. actually have that in our favor. and, where possible, hide it.

How do you help, and how do you see that helping Monica: Let me ask you, as a final question, about Jim: Right. Interestingly enough, we’ve done in terms of reducing the complexity, and getting a what it is that you at Kathrein are working on in simulations – say, for example, in Manhattan – business model that works much more smoothly order to get ready to meet the challenge over the where we’ve taken a series of conditions and than it does today? next years. parameters that we plug into design software. We compare the performance of different models. It Jim: Good question. One way is to partner with Jim: It really comes back to evolving the quality could be external antennas on light poles, versus other vendors, vendors that might focus and solutions that we have –adding new bands to Street Connect, versus a canister. We look at the specialize, say, on advertising signs, who may them for operators who really want to future- different heights of installations: how high. We already have rights-of-way in cities, who maybe proof their systems. Adding the 3.5 GHz and 5.8 have found that that’s a critical element. How high have already worked with city governments to GHz, that’s a big deal. do you put that antenna? establish agreements where they have expedited zoning permits and approvals, and in many cases, Adding the 4x4 MIMO capability, particularly at We can make evaluations, then, on which are the where they already have power and fiber in the the high-band frequencies, is a big deal. Everybody most effect tools for a given situation. It’s also ground. wants it. They all want to be ready for it when it interesting, really, that the performance doesn’t comes. Then adding these flexible products that vary that much between the Street Connect and Another is education at the government and local allow, in many cases, to install an antenna in a what we call a slim-pole antenna, which is a small levels -- providing the necessary information to nonconventional location. Those are really the antenna, or the barber-pole antenna. Actually, it’s ease the process. And another is for us to directions that we’re headed with the small cells. encouraging how well they all work in their right continually innovate with new antenna solutions place. and concealment ideas. And of course laying the groundwork and providing the latest research and development You’re absolutely correct, it is a time-consuming into 5G technologies -- especially millimeter wave. process to figure out ways to get the right-of-way,

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |64| About Kathrein Kathrein is an innovation and technology leader in today's connected world. Kathrein solutions enable people to globally communicate, access information, and use media, at home or on the road. From mobile communications, signal processing systems and optimized data transmission in buildings; to fiber optic and cable networks and satellite reception technology; to radio and TV transmission and data reception in cars; Kathrein is a hidden champion and family-owned enterprise that has been working on the technologies of tomorrow since 1919. For more information, visit: www.kathrein.com. About Jim DeKekkoek Jim DeKoekkoek took the role of Product Line Manager in April 2015 with responsibility for Base Station Antennas and Small Cells for Kathrein in North America. Mr. DeKoekkoek started his career in RF with the U.S. Air Force as a SSB Ground Radio Tech and also worked in an Electrical Installation Squadron installing aircraft control tower radios. After 6 years in the USAF, Jim held positions at General Electric Mobile Radio, Zetron Inc., and Kathrein-Scala Division. In total, Jim has more than 40 years of experience solving wireless communications problems and has a BA in Accounting from Dordt College in Sioux Center IA.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |65| advanced technologies, such as massive MIMO used to generate high-quality, digitally Profile and beamforming. modulated signals.

Rohde & Schwarz Products that help mobile operators and network FSW Signal and Spectrum Analyzer, and RTO equipment vendors test and optimize network Digital Oscilloscope, with up to 2 Gbps elements that are affected by densification include bandwidth, for RF testing in TDD and FDD Headquartered in Germany, Rohde & Schwarz has these: environments. been a main player in wireless communications testing, measurement and monitoring for over 80 Freerider III, a portable/backpack TS8980 RF Test System Family, for RF years. In addition to mobile wireless service benchmarking solution for indoor and outdoor conformance and operators’ acceptance tests. providers, it serves multiple markets, including testing. fixed wireless, broadcasting, electronics, automotive, aerospace, government, security, and TSMW Universal Radio Network Analyzer and public safety. TSME Ultracompact Drive Test Scanner, two test-drive scanners used to Its solutions encompass five product areas: test tune, install, optimize, monitor and measurement, broadcast and media, and benchmark models. The cybersecurity, secure communications, TSMW Universal Radio Network radiomonitoring and radiolocation. Analyzer is designed to support MIMO and CA. TSME In mobile wireless, Rohde & Schwarz provides a Ultracompact Drive Test wide range of products that span mobile phones Scanner is compact and hence or UEs, and mobile-network testing. also well suited for indoor environments. In terms of access technologies and services, the company supports all 2G and 3G legacy ROMES4 Drive Test Software technologies, LTE and 5G, in addition to Wi-Fi and for network analysis and short-range technologies such as Bluetooth. optimization that works across Within LTE, Rohde & Schwarz covers VoLTE and wireless interfaces. voice testing, CA, SON, Wi-Fi offload, and public safety – both for FDD and TDD. CMW500, a platform for testing the wireless interface of mobile An emerging area of focus is IoT, where Rohde & devices. It can be used across Schwarz provides measurement and testing technologies and architectures, solutions for eMTC/LTE-M and NB-IoT. In and in conjunction with SON. preparation for 5G, Rohde & Schwarz is working on products for wideband, millimeter-wave, and SMW200 Vector Signal new waveforms yet to be adopted, as well as Generator, with up to 2 GHz of internal modulation bandwidth,

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |66| address the challenges that network operators and testing, whereas before it was traditionally you Rohde & Schwarz customers, alike, face due to these new network walk out to your car, slap an antenna on, put the densification strategies. scanner in your front seat, and go do some Testing a densified outdoor drive testing. My colleague, Rob Wattenberg, is our North American Business Director for SwissQual There’s still a little bit of that, but really things are network Products. He talks with our R&D department in moving indoors. As densification methods like A conversation with Jeremy Switzerland on a consistent basis, to make sure all small cells and LTE unlicensed bands start to get of our strategic decisions are made according to used, and you’ve got Wi-Fi offload going on, the Cline, North America Product what we see in the market, and to ensure that we focus on QoE is just going to become more and Manager, and Rob Wattenburg, have a coherent product offering. more important, to reflect what customers are actually seeing and feeling. North America SwissQual Monica: There have been changes in the way we Business Director test network performance, driven by operators’ Again, I’d say it’s more about the actual data need to capture QoE. throughputs that can be achieved and other subjective measures that everyone can relate to. Monica Paolini: Our conversation on densification With densification, how is testing going to be at Rohde & Schwarz is with Jeremy Cline, the affected by moving the focus towards QoE? Monica: When it comes to utilization, you North America product manager, and with Rob mentioned indoor access. We know that most of Wattenburg, North America SwissQual Business Jeremy: There absolutely is a shift towards testing the subscribers and most of the traffic comes from Director. QoE. That’s one of our top priorities. We’ve known indoor locations. As you say, you still need to go for a while that KPIs only tell a part of the story. around with a car, but you also need to go inside Jeremy, can you tell us what Rohde & Schwarz is When’s the last time you went to a ball game at a the buildings or try to capture the indoor doing in the field of densification, and what you stadium venue, for example, and you were performance. How do you do that? How will you and Rob are doing specifically in this area? worried about “What’s my RSRP look like?” need to change the way you work or how your solutions work, to capture that? Jeremy: Rohde & Schwartz is multifaceted. We’re, Maybe some of us RF engineers do, but the generally speaking, an RF test-equipment layperson is more worried about subjective things Jeremy: One of the trends we’re seeing is that manufacturer. Rob and I work on the field side. like audio quality and video quality. They want to small cells are really moving indoors. Generally We’re constantly working on providing our know “Is my YouTube video that I want to speaking, we’re familiar with the inner workings of customers the ability to perform outdoor drive download coming through? And if it does, is it the macro network; that’s a fairly mature tests, as well as indoor walking tests, mainly to coming through in an acceptable manner, where technology and infrastructure. But as small cells verify network coverage and quality provided by there’s not a lot of blurring and things like that?” move indoors, there are going to be quite a few the ever-increasing number of small cells and all more testing challenges to address. the other strategies you hear of as part of network There’re so many aspects of what we call the real- densification. world customer experience that it’s become truly For one, there’re going to be many more small key to test and verify the power of a mobile cells to verify coverage for. It’s as simple as that. I’m the North American Product Manager for network. Something I’ve noticed over time is that Mobile Network Test. That’s a market segment for our typical use case for our products has evolved In a well-coordinated network, you hope that the which we have a dedicated product portfolio to to a more hands-on, walking-around type of small cells that are going indoors don’t really have

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |67| any effect on what’s going on outdoors or vice One of the things that comes up is RF interference. versa. If the small-cell network is well designed, That’s something you can never ignore. As if it you’re probably not going to create those kinds of wasn’t challenging enough before, you’re going to problems on the macro side. see these LTE unlicensed spectrum bands coming into the picture. Just the sheer number of small But the new challenge we’re seeing is in the third cells that are going to be deployed on the network, dimension, meaning altitude. When you want to which could have coverage issues between get access to the signal, you walk or drive around, themselves and/or with the macro cell network and typically it’s in two dimensions. For an indoor outside of a venue. venue, though, you may have 15 or more floors, so you’re doing the same route but you’re actually You’ve also got the emergence of self-optimized moving up in altitude, and now you have to do this networks, and the idea is that that will lead to less 15 different times. It’s more time consuming. in-field optimization, but interference is simply There’re more things to take into account. The test something you have absolutely no control of in tools have to be able to keep up. some cases. But you should never ignore it. You need to be worried about that in the field. From a test equipment perspective, we have to worry about providing solutions that are able to If a base station’s receive antenna can detect a efficiently and get those issues resolved as fast as verify the coverage for each and every small cell. signal that shouldn’t be there, that’s really possible. And the changes we make in our portfolio are interference, but there’re so many small driven by all the new devices that get released interferers out there that we know lie in the wrong Monica: With small cells, it’s more than just having these days. spectrum. This may not matter with a macro smaller-footprint equipment. A lot of things network, because that signal, with its path loss and change. As you say, sources of interference might As things get more complicated in the network, we its geographic location, may not be picked up by change, but also the way we deploy them and test have to keep the testing process easy. We can’t the base station, meaning it’s not really a threat. the small cells. You might have less-skilled afford to take a step back in usability. technicians installing them and testing them. The But now when you start to look at small cells, location is different, and the networks will change. Monica: Many new dimensions are added to there’re going to be so many more deployed in the The dynamics of the network change more quickly. traditional testing, where you had only a macro network. They could be closer to these sources The operator might continually add small cells at a network and phones. Not only do you have that weren’t really a threat before but now they faster pace than they do with micro cells. different devices, in different locations. You also are, because a small-cell receive antenna may pick have small cells sharing spectrum with macro cells. them up. In some ways, you could argue that How is that going to change the way you test the They affect each other. How do you deal with that they’re more susceptible to interference. That’s network? More dynamic network, less-skilled in terms of testing? just going to add another layer of complexity. employees?

Jeremy: You want to have some coordination It’s going to be important to have interference Jeremy: That’s certainly a concern. As we go on between the small-cell and the macro-cell tools just like we do today and did in the past. In from 2G, 3G, 4G, and now we’re looking at things networks, even though early on you may not have the future it’s going to be even more important to like 5G and IoT, the network is getting more and that. locate these signals quickly and easily and more complicated. You’ve got all these strategies

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |68| related to network densification, like small cells, system have really changed the way these tools heterogeneous networks, and LTE and Wi-Fi are interacted with. coexistence. We get access to the chipsets. We get from the Testing equipment is expected to keep up. We devices all these KPIs that we need in order to look have to put in all these bells and whistles, you have at some of the things I’ve already mentioned, like to be able to verify performance, but the test data throughput, and voice quality, and video equipment cannot get any harder to use. quality. We get a whole lot more information than we used to. Speaking from my own personal experience visiting with our customers, I’ve noticed a trend Again, it all comes back to the ease of use and toward the idea that they’re going to change who allowing anybody to pick it up, turn it on, and press does what in the company, and the person holding go, and get what they need. the test equipment may not have much RF background or experience. They still need to be Monica: Ease-of-use and automation are crucial able to use the tool. for dealing with increased complexity. One source of complexity that you mentioned earlier is that Ease of use is definitely at the top of the list from a operators’ use of unlicensed bands for Wi-Fi, LTE testing equipment perspective. That’s something unlicensed, or other access technologies. How do we’re focusing on, and it’s going to be absolutely you deal with that? To test for performance in an critical to have, especially if the network gets more unlicensed band, it’s inherently more complex. complex. Rob: From our test tool point of view, there are a The other thing I’ll add is that we see a need for couple things we have to be concerned with. automation and remote monitoring, as well. You look at, again, this number of small cells that are One is obviously the RF environment. We have going to be deployed in the network. They already tools to be able to properly assess that RF are, and it’s just going to grow exponentially environment. We’ve got a good line of RF scanners through 2020 and beyond. that are software defined. We can evolve with technology. We can evolve with spectrum band Expecting a cell technician, for example, to go out deployment – which is probably even more in the field and verify all of their sites in one day is important than the RF environment, which, as almost becoming unrealistic. There’re just too Jeremy referred to earlier, is pretty robust. Once device manage the flow of information through many sites that they’re going to have to be you’ve got these things installed, they tend to the IP domain? That’s where a ton of our energy is worried about. function in the RF domain. going these days.

Instead, we may see some testing requirements The next thing we really have to be focused on is We’ve got a company called ipoque that is a where you leave a scanner or a UE at a site, and the functionality. If I’ve got a device that’s made to Rohde & Schwarz company, and we have taken you can log into it anytime, 24/7, to see what’s be able to use LTE unlicensed or maybe in their deep-packet inspection capability and going on. Advances in the Android operating aggregation between LTE and Wi-Fi, how does that

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |69| integrated it into our tools so we can watch the IP have to be able to measure what they experience. delivery mechanisms out there for voice and flow. That’s where quality of experience really becomes video. They all have strategies in terms of how the critical. I don’t care if I get a gigabit per second out bytes flow. Their emphasis is on delivering the best Most of the problems we’re seeing today are not of my mobile phone. All I care about is can I have a quality of experience to the guy who’s using the caused by RF deficiencies, but through packet video chat with my kids who are away at college? phone. Latency’s not so important there. I don’t routing. Maybe some networks are filtering Those are the things that are more critical to me. care if there’s a 10-second delay in the flow of packets that shouldn’t be filtered. Maybe they’re packets. That’s not a critical element there. not being prioritized properly. Maybe there’s I can give you hundreds of examples of real-world packet loss. These are the things that we have to issues we’ve had, where we’ll go do an Ookla When we move into, say, live video, where I’m help engineers figure out how to solve. speed test and see 30, 40 mbps on a mobile watching the Olympics on my phone, now latency phone, yet I can’t download a 500 MB file in less can come into play. Again, do I really care if that We take what is essentially the end-user tool or than four hours. There’s a problem somewhere video is coming to me three seconds delayed or device, and it’s really important for us to use that out there in the network. We’ve got to be able to five seconds delayed? The typical delay we see in particular thing. We can’t use a generic device to provide the background information that watching on a mobile phone on a live video is 5 to test these complex networks. The phone that you engineers need to understand where these 10 seconds. and I go get from a carrier is what we have to use. problems are coming from. It does matter in terms of the throughputs. If I’m I can go get a phone from Europe, bring it over, Monica: That is a challenge, because different looking at a video on demand, I can get a buffer unlock it, and use it on some of these networks applications have different requirements or filled in a high bandwidth network very quickly. here in the States, and it doesn’t perform the same generate different types of traffic. If you test the You get a spike of data coming to your phone. It way. It might not work properly. overall performance of the network, you might not plays out. As the buffer empties, another spike of be able to see the difference there. data rolls through, the buffer gets filled back up For instance, if VoLTE doesn’t work, we have to again, and everything plays out smoothly, and I’m use the carrier’s end-user device. We buy that, we Rob: Exactly. I would say this in terms of latency. a happy customer. have to integrate our software into that, and then Video is really one of those hot-button items. we have to layer on top of it functions that are We’ve seen a number of studies that show that For live video, you don’t have the luxury of essentially mimicking what you and I would do video is going to be 75% of all bytes that flow in buffering 40 seconds of video. You get much with our phones. It might be, “Hey, I’m going to go the network by 2020. Looking at video and latency, smaller chunks of data, which means you’re now surf websites.” I might go look at videos. I might do those things can go hand in hand, but in some much more susceptible to dropouts and data a VoLTE call. I might do a video chat. cases, it’s not quite so important throughputs. Latency has an impact there because the higher the latency, the larger the buffer needs These are all things that we have to do to drive the If I’m looking at a YouTube video, latency is not so to be to overcome problems with it. Lower latency devices to test these networks. The complexity in critical. If it’s a video on demand, let’s say the means I can keep my buffer sizes smaller and not this densification is the aggregation of technology typical use case, I would go to YouTube and watch worry about retransmission so much, as well. bands. The functionality of each has to work in a video that somebody uploaded at some time in concert with all the others. Troubleshooting the the past; that’s really nothing more than a file It’s a very complex formula that goes into latency, problems that arise from that is really critical. download in disguise. buffer sizes, how close to real time can we get with the video or audio transmission if we need it in Just to take a quick step back, when we program In this case we’re talking about YouTube. It could real time. these phones to do what our end users do, we be Hulu, Netflix, Pandora. There’s all kinds of

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |70| Monica: VoLTE and conversational video, or ViLTE, critical for carriers to be able to measure audio it later. We call this mouth-to-ear time, and it is are the next challenge. How do you test for quality, video quality, and lip-sync capabilities on super critical from a quality of experience performance? these video chat calls. perspective.

Rob: IR.94 – let’s say an IMS-type video chat – is I would say the biggest challenge for us, and we’re Nothing’s more frustrating than what we call very challenging to test in the real world. We see a very close to releasing this, is that we have to be double talk, where I talk and you talk at the same lot of people who can put things back in the IMS able to inject video into the device, because we time because we’re not quite in sync. There’s an IT servers and try to assess video quality, but in the can’t really test reliably with opening up the standard in terms of when double talk is end, can you tell if I am a happy customer? The camera feed and making a video call. We have to problematic. Latency plays into that in a big way. only way to tell if I’m happy is by being where I’m feed into the device a known reference clip that’s at with the device. going to simulate the network properly. That’s a real problem area for carriers. In terms of VoLTE and ViLTE, that’s probably one of the more When I talk to my kids, for instance, we’ll pull up A black screen’s not going to transmit any bytes, critical parameters that we’re able to measure. Of the phone, we’ll do a video chat call. What but motions, face movement, those are all things course, you still have to look at things like jitter, happens behind the scenes is everything’s getting that actually up the bit rate of the data jitter buffers, the call state that you happen to be routed through the IMS servers. There’s a video transmission. We have a video/audio clip that goes in. socket set up for video packets. There’s another out at the same time, and then we put special socket set up for audio packets. markers in the video so that we can now calculate There’re some specialized tests that we have, just timings for the lip sync. to cover where you’re at with that. We have a What we’re doing with our test tool is we’re now special test that we can run to measure the exact running what we call an algorithm. By “algorithm” Then, on the receive side, we run these algorithms, amount of latency you have. It’s done through a we mean that we can run audio into an algorithm, as well. Those receive algorithms, we’ll be able to pretty clever mechanism where we send audio and what’s going to get spit out is a mean opinion go full-duplex audio, video, and quality. On the from what we call the A side to the B side. The B score that’s going to grade the quality of that algorithms, these aren’t simply software pieces side hears it and sends it back, and we can get audio from one to five, five being best. I can grade that just grade this video. These are algorithms latency measurements from an end-user the audio with this algorithm, and I can grade the that have gone through incredible amounts of perspective based on that. video with a separate algorithm. Now I’ve got two testing. measurements, or two KPIs, if you will. We also have the ability to layer noise in on top of The ITU has issued standards for the proper way to speech so we can see how well the network The third thing that’s critical to understand in a measure video quality in mobile environments. elements that remove unwanted noise are video chat is lip sync. Since audio and video are ETSI has recommendations as well. We work very performing. If we can throw background noise in, transmitted on separate paths, you’ve got to be closely with the standardization bodies, and we’re say, the conversation we’re having now, hopefully able to come up with a method for calculating lip very proud of the fact that we have the approved the other end doesn’t hear it. That requires all the sync. ITU standard for measuring video quality in a right noise cancellation mechanisms to be turned mobile environment. It measures all the way up to on. We’ve seen some studies where they actually put 1080p, and we’re in the process of testing it for 4K blood pressure monitors on people, and in a poor video, as well. Echo is another problem. We’ve got to make sure video environment, blood pressure actually goes echo cancelers are working properly. We have up in people. We know that we don’t want to be If you’ve got real latency issues, you’re going to specialized tests just to simulate echo. We can under any more stress than we already are, so it’s end up with real problems of when I talk, you hear introduce echo and measure the ability of the

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |71| echo cancellation components to remove that. ourselves prepared for running the kinds of tests you see behind me create interference that cuts Again, it all points back to this quality of that are going to stress these new networks and throughput of an LTE network in half. experience, measuring audio and video quality. their functionalities. That’s one of our challenges in looking forward. Finding those kinds of sources of interference is Densification reemphasizes the need to look at this going to be one of those really tricky things for the in this manner. We’re in this heavily densified How will the internet of things matter to us on the carriers. It’s spotty. Interference could be up and network. Let’s take ourselves forward 10 years. testing side? If my doorbell is going to connect to down, all over the place. Coming up with a We will have devices that may be hitting the the internet, I don’t think we need a test tool to mechanism to do that more efficiently is Internet or hitting the IP backbone in several ways tell us does that thing work properly, but we do something we’re working on. in parallel. It is reemphasizing the back-end have to understand what its impact on the functionality, and how it might impact the quality network is. Back to your point earlier, which is everything has of experience. to be simpler than it is today: we have the need for Network virtualization is coming, and mobile more and more people hanging networks out in Monica: Absolutely. If we look at the next 10 devices might have some clever ways of accessing the world, the real world, whether it’s in small years, what are the challenges that you are trying multiple radio-access technologies, or share RANs. buildings. to address with IoT or 5G? How are they going to All these have an impact on functionality, so we’ve change the way we test networks? got to be able to stress that functionality. Those I’ve heard some people talk about moving towards are things we’re working on, as well. 100,000 sq ft buildings and bigger. That’s where Rob: From a tools perspective, we tend to sit back we need to be focused. It’s going to go to 50,000 and wait to see where the markets go, where the The RF environment is going to get more and more sq ft buildings and then maybe my house. carriers are actually deploying technology, and complex as 5G comes. Our RF scanning receivers, then we’re ready to test it. which have these phenomenal performance That’s going to require somebody to do some kind capabilities, and software-defined radios, we’re of work on site. That guy that does that work is The nice thing about Rohde & Schwarz is we’ve got now evolving those to make sure we can really going to have to be extremely efficient. He’s going groups in this company that work with the chipset understand the RF environment. to have to be able to go deploy, hook up, and test. manufacturers, the device manufacturers, the No longer will we have the days of engineers infrastructure manufacturers, the standardization One thing, and Jeremy talked about it earlier, is we coming out and doing surveys after the fact. bodies. We are in the ones and the zeroes of are looking at how we can better assess evolving technology. interference in this evolving market. Right now, we The holy grail for us is a tool that’s absolutely for feel LTE is very susceptible to interference. We’ve free. You hit start, stop, pass, fail, and everybody’s Here on the field side, on the test tool side, we’re seen things like ballasts and fluorescent lights like happy. Obviously we can’t get to the holy grail, but leveraging that experience, and we’re getting that’s the direction we’re heading.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |72| About Rohde & Schwarz For more than 80 years, Rohde & Schwarz has stood for quality, precision and innovation in all fields of wireless communications. The electronics group is strategically based on five pillars: test and measurement, broadcast and media, cybersecurity, secure communications, radiomonitoring and radiolocation. The company addresses customers in the mobile radio, wireless, broadcasting, electronics and automotive industries, in aerospace and defense as well as government, security and critical infrastructures. Rohde & Schwarz is among the world market leaders in its established business fields. It is the world’s leading manufacturer of wireless communications and EMC test and measurement equipment, as well as of broadcasting and T&M equipment for digital terrestrial television. About Jeremy Cline Jeremy Cline is a Product Manager at Rohde & Schwarz where he is responsible for products that are primarily used for mobile network testing. During his 6 years with the company, he has specialized in drive test, benchmarking, network optimization, interference hunting, and other general purpose RF applications. Jeremy helps identify testing solutions for RF customers across multiple industries including aerospace and defense, telecommunications, semiconductor, biotechnology and device manufacturing. Jeremy graduated from The University of Texas at Austin with a BSEE, and The University of Southern California with a MSMDDE. He can be reached via e-mail at [email protected].

About Rob Wattenberg Rob Wattenberg is the Business Director for SwissQual products in North America and is based in Irvine, California. Rob joined SwissQual in 2008 as a Regional Manager, and introduced their innovative products and technologies. He brings over 25 years of experience to Rohde and Schwarz in the Mobile Network Test market segment. Rob has held several executive level positions in the wireless industry; primarily related to field testing, optimization, and benchmarking of cellular networks. Rob’s expertise in recent years focuses on Quality of Experience measurement as it applies to voice and video. He holds a Bachelor’s degree in Electrical Engineering from Cal Poly San Luis Obispo.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |73| policy. They support both data and voice enables operators to manage interference among Profile (VoLTE) for eight concurrent active users, with small cells and between macro-cell and small-cell up to 225 mbps throughput and a coverage layers. It uses advanced scheduling and cell- Samsung Networks area up to 1,000 sq m according to Samsung. coordination technologies that benefit from the tight synchronization that centralized baseband Samsung is also working on LTE Unlicensed small processing allows. Samsung Networks develops network cells, with the first products supporting LTE-U – the infrastructure products for mobile operators Samsung LTE-U eFemto – using enhanced Carrier Current R&D focus is on 5G: Samsung is working worldwide. Since the 1970s, Samsung has been a Sensing Adaptive Transmission (eCSAT) for on the new RAN architecture (virtualization, with leading global mobile infrastructure vendor, and coexistence with Wi-Fi. The LTE-U small cells will C-RAN topologies and decentralized core functions more recently an innovator in TDD-LTE and FDD- be software-upgradable to LAA. moving toward the edge); new bands (mmW LTE technologies. bands: 28 GHz, 37 GHz, 39 GHz, and 60 GHz, Samsung’s small cell gateway, helps operators primarily); improvement of antenna technologies Samsung Networks launched its first commercial increase their densification efforts, by aggregating (e.g., MIMO and beamforming); and coexistence FDD-LTE network in the United States in 2010. This thousands of small cells in to a single connection to with legacy RATs (3G, LTE) that are still widely was followed by commercial TDD-LTE network the EPC. Samsung’s end-to-end small-cell network used. launch in 2011 with worldwide deployments from Asia to the United States.

A core focus of Samsung’s technology development has been on an end-to-end small-cell portfolio featuring three types of small cells:

Outdoor small cells, with radio and baseband combined in a compact unit that can be installed on outdoor urban fixtures to increase capacity in dense areas, and to improve coverage where the macro layers cannot reach. According to Samsung, its outdoor small cells can support up to 200 active users in a 20 MHz channel. Indoor small cells, suited for installation in public venues and enterprises. These are smaller than outdoor small cells (4 kg) and support 64 active users per cell. Residential femto cells, small units (less than 1 kg) designed to be plug-and-play devices that automatically select the appropriate frequency and enforce the operator’s location

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |74| mind we said, “OK. What is the best way to on the cell edge of the macro network – they are Samsung address a need within the United States?” the ones who consume the most power and spectral resources, from a macro standpoint. Most Densification from As part of that, we started to look into these in-building consumption could also be classified as miniature base stations. That eventually led to our cell edge users due to RF signal characteristics. Our world’s-first small cell – at that time we were customers asked “OK. Can we also use this product the chipset to small calling it “Internet Radio” – launched with Sprint. to see whether it could be utilized as a capacity cells That was back in 2007. offload tool?” That was all part of the 3G. A conversation with Nivi At Samsung, my role encompasses responsibility As part of evolution from 3G to 4G, now the for radio access solutions targeted for the United biggest need we see is voice over LTE. “Can I Thadasina, Senior Director of 5G States. We do the feasibility analysis, propose new provide voice coverage in areas that are very hard and 4G Engineering, Samsung ideas and methodologies, new product to reach?” For that, our customers are looking at requirements and the product definition. Once our small cells as “Hey, can I use this tool to address Electronics America development team in headquarters realizes the VoLTE coverage?” product, it is validated here in the United States before the final delivery to our customers. Our With that, obviously, we can provide a better user Monica Paolini: For this conversation on team is also responsible for commercial rollout of experience – better voice quality and significant densification, I have the pleasure of talking with small cells for both enterprise and residential improvement in call accessibility and retainability. Nivi Thadasina, Senior Director of 4G and 5G markets. Another added benefit for operators is Engineering in the Wireless Network Systems improvement in spectral efficiency for in-building division at Samsung Electronics America. Monica: Samsung has been involved in small cells utilization and cell edge users. We are able to for a long time. I remember in 2007, the Sprint squeeze more bits per hertz while improving Nivi, may I ask you what Samsung is doing to help launch. A lot has changed since then, even though overall experience for end users on their cell densification efforts by mobile operators? And we’re still in the beginning of the densification phones. what is your personal role within that? process. Can you tell us how you see the evolution to densified networks? Looking ahead, as you’re aware, 5G is the next Nivi: I’ve been with Samsung Networks for over 10 evolutionary step. Millimeter wave can have a very years and have been involved with small cells since Nivi: In 2007 the need was coverage. In-building large path loss. Propagation characteristics are their ideation and inception in 2004, even before was the challenge. The signal was not penetrating tough. I would say that the 5G is a very natural the term “small cells” was introduced. We did the where it needed to. The immediate need from our evolution for small cells. initial research – partly because of Samsung’s customers was “Can we get coverage in areas capabilities to undertake development of new where it’s very hard to reach through macro Monica: What you were saying with respect to the technologies and to solve tough problems faced by cells?” Coverage was the initial need. increased spectral efficiency in 4G is interesting. the operators. What you’re seeing is an increase in spectral With the introduction of small cells they now had a efficiency in the macro. For instance, indoor small We design and offer end-to-end products, as you tool to quickly and efficient address coverage cells not only improve the in-building performance are aware. Plus, we have the capabilities to design challenges in their network. That eventually led to for the people inside the buildings who are closer and manufacture our own chipsets. With that in “Now that we got the coverage, can we also look to the small cell, they can also improve the macro at capacity offload?” Especially for users who are layer. Is that correct?

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |75| Small cells help the macro, as well. Usually in the beginning we thought, “You just put a small cell for the people using the small cell.” It may benefit even more the people who are at the edge of the macro.

Nivi: Absolutely. I hear many encouraging comments from the operators about average sector throughput increasing after addressing those tough users deep inside a building or on the cell edge.

That’s mostly due to the fact that you are able to close the link between the UE and base station with improved MCS rates, higher bits per hertz and less power thus resulting in lowered interference levels. Definitely, I think it’s a win and win for both deploying a small cell deep inside a building, they problem is with their home Wi-Fi coverage or the the macro-cell and small-cell layers. are now able to light up the high valued resource. operator’s macro network. The operator is unable to trace user experience since that particular call is Monica: Do you see a shift between indoor and From Samsung’s standpoint, we see this as a great handled by a network that is outside of their outdoor locations in the operators’ densification opportunity, as we’re able to help our customers domain, i.e., unmanaged network. efforts? to address and relieve the macro network by reducing capacity demands from users deep inside One way to address it would be by moving that call Nivi: Samsung has been focused in indoor for a a building. Plus, you are able to go address the from Wi-Fi to voice over LTE, all of a sudden I am couple of different reasons. voice over LTE challenges that operators face, now able to go see and predict the user which are typically in indoor locations. experience. That’s one thing. There are already widely published statistics showing that 50% to 70% of the calls originate Monica: Couldn’t the operators use Wi-Fi Calling The second is that it’s well publicized that with from deep inside the building. instead? Wi-Fi, the coverage within your home is still somewhat restricted compared to coverage you There are also estimates on how much of the Nivi: Very good question. There are different can experience with the small cells. Based on our indoor spectrum is under-utilized by the operators. opinions on that. Let’s say you make a voice call product experience, we have received positive These numbers vary; however, the primary over Wi-Fi and that the call experiences voice feedback from the field on how noticeable was the concern is the disparity in coverage between the quality challenges. Or, worst, the call drops. coverage expansion in their homes relative to outdoor and indoor environments. Transitioning or handover between Wi-Fi and Wi-Fi coverage. Now they are able to make calls in macro networks is not seamless, and may result in areas where they could not before. Spectrum, among others, is clearly the most poor voice quality or frequent drops. In that case valued asset owned by an operator and that is what would an end-user do? The user will Monica: 5G is a huge opportunity, because under-utilized in indoor environments. By naturally call the operator to complain since most wireless networks will be able to use a wider range end-users can’t decipher whether the root of the of spectrum bands.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |76| At Samsung, how do you see those new bands? location because it’ll have less interference. At the vendors, but integration is not easy. What’s your Some of them are traditionally not cellular bands, same time, you need more equipment because view on this? like millimeter wave. How do you see them being the range is smaller. used for densification purposes? Nivi: When we first approached small cells, we If you have a cell with a range so short that you encountered similar challenges, with not only 3G Nivi: You saw the FCC ruling that came out need a high number of them, is it cost effective for but also 4G networks. Typically, most operator recently. They’re immediately looking at four an operator? networks for macro deployments are homogenous bands. That’s great news for all of us. The 28 GHz, within a given market. A single vendor covers the 37 GHz, 39 GHz, and 60 GHz. Right now we’re Nivi: We are looking into various technologies to entire network. If you are not an incumbent doing a lot of work in 5G in this area. We’ve been extend the range of 5G in millimeter waves well vendor, this can introduce additional challenges. looking at 5G for the past several years. beyond what the industry initially envisioned. Does Knowing that we are not an incumbent player in all this mean a millimeter wave based 5G cell will the US markets, we designed our products from From our standpoint, we see a great opportunity. equal sub-2 GHz cellular levels? Not yet. day one to underlay a macro network. Intrinsic to There is an immediate opportunity in 28 GHz. the base layer of our HW and SW is ability to co- Overall we are looking at multiple different Combination of massive MIMO and adaptive beam exist within any incumbent markets. solutions for 5G to provide blanket coverage in formation, as an example, will significantly both outdoor and indoor environments. By virtue improve the link budget by virtue of generating Now the question is “How can we coexist?” Not of Samsung’s strong presence in small cells, we narrow beams and large improvements in only can I coexist, the end user should not know strongly believe that we have a distinct advantage transmission gain. I think some of these the difference that “Hey, I’m on a Samsung as we have deployed the most field proven technologies and techniques are utilized to network or someone else’s network.” The handset solutions in this space. overcome the path loss. should seamlessly move between these multi- vendor solutions. Monica: What kind of bands are you seeing that We do believe 5G is ready for commercialization. are better suited for indoors versus outdoors? Is it We have done extensive demonstrations and trials There are certain innovative ideas we had to backhaul or access? What’s the best way to to validate how to close a 5G link at much greater incorporate into our products to seamlessly leverage it? distance to allow cellular like deployments. coexist. On top of that the coexistence will come in two forms. One is “How can you coexist?” from a Nivi: Our customers will deploy all the way from I do believe that the 5G deployment pattern will radio standpoint, and the other is “How can you sub-1 GHz, namely 600 MHz, to millimeter wave. be combination of macro cellular and small cell coexist?” from a core network standpoint. There It’s already well known that once you reach footprint. It will utilize various configurations from are different solutions to address the radio and the millimeter wave, the in-building penetration can street furniture, rooftops, and existing cell towers core network. be a challenge. We have product solutions that to fulfill coverage and capacity demands. allow our customers to solve such problems. Our In my opinion, 3GPP has done a very good job in customers use various high and low bands to Monica: There are going to be many solutions out terms of coming up with well-defined interface selectively address various use cases and coverage there. That brings up the issue that not only do specifications, which make it bit easier for multiple needs inside and outside buildings for access. you have different solutions, different locations, vendors to interop. However, there are certain but also different vendors. How do you view that? things operators can still do, especially in the X2 Monica: There are two different views you can This has been a challenge for a lot of operators. area, to further promote multi-vendor co- take. As we move higher in frequency, the range is They want to have equipment from different existence. shorter. That can be an advantage in an indoor

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |77| SON is another area in which interoperability is crucial. It’s not necessary to have a vendor- agnostic SON. But, certainly, having that would help promote a truly multi-vendor radio network.

Monica: There are also different technologies. We have LTE, but operators are also using Wi-Fi as a complementary radio access channel. Then we have LTE unlicensed – either LTE-U or LAA, LWA, MulteFire. This opens the way to another level of tight integration of performance across different access channels. What are you doing there? How do you see this further developing?

Nivi: You’re right. The technology is getting more and more complicated. Not only are you looking at multiple bands, multiple technologies. Now you’re looking at a combination of LTE and Wi-Fi and so on.

There might be three or four different flavors of this technology. You might see a combination of LTE and Wi-Fi, since Wi-Fi is prevalent. You might see LWA solutions out there. LTE-U is something that we are aggressively pursuing. Certainly you’ll see that plus LAA. Of course, MulteFire is also out there. 4G into 5G. Plus, as an industry, I think there is a plays a strong role and hopefully influences the rapid evolution from physical elements to evolutionary path for our customers. Each of these solutions has its own pros and cons. virtualization. Eventually this industry will evolve and hopefully Monica: You have your own chipsets. You will converge. Samsung is clearly looking at all the different mentioned virtualization. Does it mean also that opportunities, technology evolutions, talking to you’re working on the C-RAN part of the solution Monica: What are you working on with Samsung different customers, not only in the US but for small cells? today in preparation for the next five years? What worldwide, in how best to design our solutions to is the focus for the future with respect to fit their needs. Nivi: We have launched C-RAN in Korea. I think densification? that’s already well publicized. One of the biggest strengths Samsung has is that Nivi: We are certainly working on lot of exciting not only can we provide complete end-to-end When I mentioned virtualization, the approach to things, in terms of how we can capitalize on our solutions, but also build our own chipsets. That virtualization for core and RAN will differ. Unlike experience and lessons learned from 2G, 3G, and

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |78| for core networks, I do believe that RAN essentially either a complete eNodeB in one box As we evolve towards 5G, what makes sense is a virtualization will probably encompass with, say, an Ethernet backhaul or low power RRH combination from having all the layers stacked in combination of physical and virtual elements with each to scale to meet market goals. one box, to a combination of having a radio plus various functional-layer splits. A small cell is everything else perhaps virtualized.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |79| About Samsung Headquartered in Ridgefield Park, N.J., Samsung Electronics America, Inc. (SEA), is a recognized innovative leader in consumer electronics, mobile devices and enterprise solutions. A wholly owned subsidiary of Samsung Electronics Co., Ltd., SEA is pushing beyond the limits of today’s technology and providing consumers and organizations with a portfolio of groundbreaking products in mobility, virtual reality, wireless infrastructure, wearables, electronics, and home appliances. Samsung is a pioneering leader in smartphones and HDTVs in the U.S. and one of America’s fastest growing home appliance brands. To discover more about Samsung, please visit www.samsung.com. For the latest Samsung news, please visit news.samsung.com/us and follow us @SamsungNewsUS. About Nivi Thadasina Nivi Thadasina is Senior Director of 5G and 4G Engineering at Samsung Electronics America, where he is responsible for radio access network products in the U.S. Previously, Nivi led engineering efforts for the first commercial launch of LTE service in the US market with MetroPCS that utilized Samsung’s evolved NodeB and evolved core packet products. Nivi also led the development of the world’s first commercialized femtocell and holds several patents for radio frequency technologies optimized for indoor environments. Prior to Samsung, Nivi worked in engineering roles at Bell Northern Research and STMicroelectronics, where he was involved in the design of 2G and 3G Wireless communications systems and Very-large-scale integration (VLSI) chipset development. Nivi holds a Bachelor of Science in Electrical Engineering from University of North Carolina, Charlotte, and a Master’s Degree in Electrical Engineering from the University of Texas, Dallas.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |80| Profile The ALLIANCE platform also includes: both fronthaul and backhaul. The ALLIANCE BIU and eBIU, the head-end The DWDM solution, Infinity Access, supports SOLiD that filters traffic to the base stations. multiple protocols (e.g., CPRI, OBSAI and Ethernet) and can simultaneously support multiple access The ALLIANCE DMS, used to manage the DAS. technologies (e.g., LTE and Wi-Fi) in the same Over the years, SOLiD has developed RF amplifiers, strand. Operators no longer need to add a new RF radios, and optical transport solutions designed The ALLIANCE OEU, an optical multiplexing fiber link when they add a new RRH or a small cell. to enable cost-effective densification of mobile device to expand coverage to additional Moreover, because DWDM allows operators to networks, increasing capacity and coverage buildings. gradually add new links to the same strand, they through DAS and small-cell deployments. can reduce their deployment and operating costs SOLiD also offers the EXPRESS Single-Carrier DAS, a as they expand their networks. SOLiD offers both outdoor and indoor DAS multi-band solution developed for indoor and installations that target a wide range of locations, outdoor environments where only one operator from large venues with high capacity-density has a presence. requirements, to slightly smaller commercial venues – which SOLiD calls the Middleprise. In the The EXPRESS Public- past, Middleprise venues have been challenging Safety DAS is a for DAS deployments because of the complexity variation of the Single- and cost of traditional DAS. SOLiD’s goal is to Carrier solution, change this and make DAS attractive to smaller designed for the venues as well as to large ones. public-safety market.

SOLiD DAS solutions support multiple verticals, In the optical backhaul including entertainment (e.g., stadiums, arenas), and fronthaul area, mass transit, healthcare, education, retail, SOLiD has been a hospitality and enterprise. leading proponent of DWDM, which splits a The ALLIANCE Multi-Carrier DAS platform is single fiber strand into SOLiD’s flagship DAS solution. It supports neutral- multiple bidirectional host deployment models, and works in channels to increase frequencies ranging from 150 MHz to 3 GHz. the capacity of the Remote units are available at different power fiber, multiplying the levels, each suited to a different venue and capacity of the link. topology. Lower-power ROUs are well suited for Linear add/drops smaller venues and public safety networks. Higher- enable the operator to power ROUs are most commonly deployed use the same strand to outdoors or in venues with high traffic loads. serve multiple small cells, and to provide

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |81| Ken: There’s no doubt we’re more focused on in- still something you have to consider. You either SOLiD building than on outdoor, mostly because we see deploy on a different channel, or you have to deal there’s a huge demand coming from the in- with interference. Densification in the building space. We all know that. In terms of the time spent, that’s also where more of the Ken: Until we have all those additional bands to problems are, so to solve those problems would play with and the use of unlicensed bands with Middleprise be very striking. LTE-U, there is a need for systems and software A conversation with Ken that are getting smarter. Most of what’s going on in the outdoor space right Sandfeld, Executive President of now is small cells – deploying more and different However, there may be a period where the in- SOLiD America types of small cells and solving for that building handoff to the outdoor, and vice versa, environment. So we’re also working on what will not happen. I do believe there is just going to would be essentially outdoor DAS solutions to be a gap in the data as you go from inside to Monica Paolini: Today I have the pleasure of compete with small cells, or augment and work outside, maybe only a second long, but there will talking with Ken Sandfeld, Executive President of with small cells. There are some significant oDAS be some issues. SOLiD America. solution advantages over small cells: oDAS can be smaller, and deliver more service and more Over time it will get better, when you have Ken, let’s start by talking about what SOLiD is spectrum than a small-cell solution can. It just stronger virtualization of software-defined doing in the area of densification, and what you’re hasn’t evolved to that point yet, but we think it networks. You’ll essentially have APIs that will personally involved with. really can. We’ll have to see how that goes. allow an ecosystem of people to all play on the same playing field when it comes to SON, and Ken: At SOLiD, we’re focused on in-building Monica: How about the cost? being able to optimize third-party solutions. Once wireless coverage with innovative indoor DAS that happens and solutions are more standardized, solutions and bringing new in-building Ken: The cost is actually lower, because you’re then we’ll see a lot more cohesiveness in those technologies to the enterprise – and now putting less electronics on the pole. The largest hand-offs. specifically to the Middleprise segment of the part of the electronics is kept back at the C-RAN business, to be able to broaden that market space. location. From a future-proof perspective, you are Monica: We had this great promise of having Middleprise densification is a priority right now. actually not pushing protocol-dependent massive densification on day one, of small cells electronics out to the pole, and that fact has everywhere, but it’s actually taking longer. Why do The other area that we spend a lot of time in is always made DAS so attractive from an outdoor you think that is? new solutions for outdoor environments – in perspective. particular for dense urban environments. We Ken: From an in-building perspective, the believe we can do that differently than some of It doesn’t mean you’re giving up a lot of challenge with the solutions currently on the the solutions that are currently available. functionality. You still have all the controls at the market is that either you have a solution that has C-RAN to be able to do SON and interference more active electronics, where you’re putting Monica: How is the balance between indoor and management. You’ll just be doing that at a radios on the ceiling, essentially a more traditional outdoor infrastructure changing or evolving with different location. approach, where you’re putting a small cell on the time? front end of a DAS. Monica: If you have indoor infrastructure, you are more shielded in terms of interference, but that is

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |82| Alternatively, you may simply deploy a small cell in If the enterprise can’t connect to the service, Ken: They don’t need to lose control or exclusivity a closet and then just radiate the signal from the including the operator connection portion, then it at all. In the model that we see developing, the closet. doesn’t matter how good the RF solution is, it’s technology allows every operator to have its useless. You ultimately need one simplified bands, its radio resources, and its management To date there hasn’t been a hybrid solution that solution that will bring those two things together. capability completely separate. brings it all together. Frankly, it’s what the network operators require; also, carriers can’t support an environment where There will be some convergence, maybe at the You have the smaller buildings where you can put they have to roll trucks and engineers to turn up a antenna, maybe at the amplifier. Of course, the two, three, or four eFEMTOs, small cells, and you small project. It needs to be a self-installed operators will want to know if they’re getting good just put them in the building and you cover that solution, and potentially self-optimizing or self- KPIs. The goal is that they shouldn’t have to do smaller building. conditioning too. anything, except if the red light goes on that says the KPIs are bad, so that someone fixes it. As you get to larger spaces, your options increase. Monica: How does the business model for small Either you distribute RF or you can distribute cells work in practice? How does it relate to the The goal is to be able to deploy tens of thousands radios around. There’re some solutions that do DAS neutral-host model? of projects, without the network operator having both. The problem with distributing radios is, how to service and maintain all of them. In order to do do you scale with multiple operators, more bands? Ken: SOLiD believes that DAS and small cells will that, we need each person in the ecosystem to be How do you deal with the additional cabling? And become a hybrid, yet simplified, solution and will able to take responsibility, and manage what’s deal with the costs associated with adding more to be deployed by third-party owners and/or neutral appropriate. That’s going to be done through a it? Your TCO goes up. Or you can distribute RF. hosts, as they often work directly with building centralized cloud capability and lots of owners that don’t want to own, manage or deal virtualization. Monica: There are different solutions out there. Is with these systems. it the technology or the cost that has been slowing It’s a one-box, software-defined, single system, down deployments? This will change the business model for the where there’s no need, as you scale, to rip and operators. Operators don’t want to devote a lot of replace infrastructure. That is a huge thing for the Ken: There are two parts to that. resources. They just want to manage their service enterprise. Anything that’s obtrusive and requires and their KPIs to guarantee that their licensed rip-and-replace is really a deal breaker. Both the There is the actual infrastructure of distributing or signal is not being interfered with. enterprise and operators know it has to happen as radiating the RF, delivering the RF to the user. I technology evolves, which is why DAS has always think those solutions need to be Wi-Fi like, low- Ultimately, all the parties need a win-win solution. been very attractive. cost devices that can support multiple operators. In order to do that, you need to bring the whole Ideally, you want solutions that can scale to thing together in one simplified architecture. The DAS is essentially a dumb pipe, so there’s a part of multiple protocols and multiple bands without days of having multiple, disparate systems – it’s DAS that’s really attractive to an enterprise. There having to add more electronics on the ceiling. just not affordable or attractive for anyone to is a part of DAS that’s very unattractive: the cost of deploy that. the cabling, and the ever-challenging service or Then you have the connection to the core, which is core connection. probably the bigger cost driver, because that gets Monica: When sharing infrastructure, operators into the business model issue. may worry that they will lose control and Monica: One issue that comes up all the time is exclusivity. How do you think they’re dealing with who’s going to pay for this infrastructure – the that? equipment and installation?

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |83| Ken: I think you’re going to see all models continue the carrier coordination, or at least to get approval these systems, as soon as they’re more capable to be implemented; however, the one model that to turn on their signal. Those companies may ask a and the business models are ready. you’re going to see become more prevalent is the third-party system integrator or contractor that enterprise spend. The only reason the enterprise, has an existing working relationship with an Monica: This reflects the increasing needs of the and specifically the Middleprise, doesn’t spend on operator. enterprise. The enterprise is willing to pay, these solutions now is there’s no guarantee, or because it has bigger needs. there’s no path towards easy connection to the The players will stay very similar, but we’re going carriers. to be able to engage projects that just weren’t Ken: In the Middleprise, lack of voice coverage was feasible in the past, and that’s really the goal, a big problem. Today it is still an issue for many There’ll be some projects where the carrier will opening up the Middleprise market. When this venues and properties. decide that it wants to put capital in, because it happens, the ecosystem will have to evolve to has an interest in that project. meet Middleprise requirements. The operators However, many things are migrating towards will also evolve, because they’re looking to grow database architectures, and so now it’s more of a Third-party owners will continue to grow in this their services and revenue base from the users. mobility play. As PBXs start to go away, enterprises space. Their ROIs will change how they get the see the need to enable everyone’s technology and revenue and how they operate. And you are Additionally, system integrators will expand their everyone’s service provider. correct in asking who pays for all of it. It can be 3PL offer to the Middleprise, and they too will get systems or carriers. Maybe it’s an advertising smarter about process and their options to deploy model. Or it’s that the building owner’s participating, and it’s managing the system.

It is going to be a somewhat different model than the typical third-party projects, but all those models will exist, and the technology will just allow operators to bring the costs down, and they’ll be able to tackle projects that are smaller than the typical large venue.

Monica: If the enterprise pays for the installation and equipment, it has a stronger position when negotiating with operators. The relationship between the enterprise and the operators is bound to change.

Ken: Yes. Some enterprise customers are going to be large enough that they will have someone on staff who will manage some of that.

However, there is a large portion of the Middleprise that will need someone else to handle

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |84| Middleprise customers tell us they would like all Ken: We believe that the more we can put in the communications, because it is a real problem, one operators to have the opportunity to become part cloud, the better. The more we can provide to the that SOLiD recognizes as a public imperative. of the solution. If they have to choose, they’ll pick network operator from our cloud at a neutral the bigger ones, but ultimately they don’t want to location, the better. The less often a network I believe the in-building systems and the systems leave out anyone. operator has to roll a truck to a customer’s site, go that SOLiD produces will ultimately be able to into the customer’s building and work on things in solve for both of those needs. They may be two The system not only needs to be ready, out of the the building, the better. Never having to physically different systems. They may use separate cables, box, to have all operators connect, but all the go to the customer’s site is the ultimate goal. but they can be managed by one cloud. operators need to understand that they must be willing to communicate and work with those Network virtualization and SDN are critical. You Regardless of how they go together, the enterprise smaller projects. have an intermediary that is aggregating and is going to require that the provider of the solution providing that service to the network operator so be able to put it all together as one system Monica: Do you think mobile operators are getting operator can aggregate tens of thousands of regardless of what the code requires for certain more sensitive and more open to the needs of the projects all over the US without having to be able cables and certain systems. Middleprise? to manage every one of those properties. In- building public safety codes are rapidly Ken: Absolutely. Network operators recognize that It becomes a much more manageable proposition becoming requirements, albeit almost randomly they need solutions, so they need technology. to network operators and becomes very valuable, across the country and at varying levels of They recognize they need business plans to be but they haven’t been able to do that up until requirements based on local authority code able to provide better in-building service. now. adoption plans. Any system in the building needs to be able to support those public safety bands in The network operators are absolutely clamoring It is only in the last few years that we’re getting to the future. It can’t continue to be two separate for better solutions to allow them keep their the point where network operators are able to systems from a management and an overall Middleprise customers happy. If that means they work with a multitude of software vendors for the maintenance perspective. have to supply a system where the other virtualization of their network, and able to add operators potentially can connect, that’s fine. It’s capabilities and open up their core to new models When you think about it, wireless communication going to be a question of who’s first. and new ways of doing business. That is the is safety. A 911 call is a voice call, but there’s also enabler. data associated with it – critical first-person In that respect, operators will be able to expand information regarding the emergency, plus there is their Middleprise presence. They may not provide Monica: Another thing I want to ask you about is location technology. capital, but they may provide backhaul, or better public safety and indoor coverage. customer service in that particular project. They’re By the way, enterprises nowadays require that the going to be fighting for their piece of that Ken: It is something that’s often forgotten. You 911 system work to code and be compliant, but Middleprise. have two schools of thought out there. You have they also are looking for location. They want to be those who are focused on the commercial cellular able to know where the people are in the building Monica: How important is it to get virtualization in side of in-building, and the fact that the public for safety purposes. the access network? And what will we put in the relies on their smartphones to ensure their safety. cloud, and what will be distributed to the edge Then you have a group of people who focus on Depending on where people are, location is a huge instead? public safety, specifically the first responder issue in public safety, as well as a potential revenue producer because of the need to know

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |85| where your customers are in a building. There’s a We don’t know exactly how the protocol’s going to network operator’s going to be looking for ways to lot of opportunity for location – but also a lot of work and how that’s going to function, but it’s make that happen. undiscussed privacy issues that have to be going to be very micro-hotspot focused when resolved. you’re talking about those millimeter-wave Ideally I think network operators are going to get technologies. more involved with supplying pipes to the building, From a public safety perspective, knowing the and dare I say, there’s going to come a time when location of the device is a no-brainer. If there’s a Ideally, you’d like to work in lower-frequency potentially an enterprise deploys a large system device in the building, it may not have a person bands so you can deal with better propagation, that simply connects to that pipe. with it, but we need to know where all the devices but the millimeter-wave-frequency bands of 5G are. The great thing about smartphones is that are going to be absolutely critical. Essentially, you may get to a quasi-roaming people rarely let them go. situation, where if the building owner is paying for For example, in places like hospitals, where you the pipe and a device comes into the building and Monica: How are we going to change the way we have extreme data requirements, you’re going to it uses a lot of high-speed data in those micro- densify networks in the future? see micro-hotspots, where you’re going to utilize hotspots, there may be a back-charge agreement those frequencies on a very localized basis to solve that compensates the in-building network owner Ken: High-frequency bands in 5G are going to high-capacity data needs. You’re not going to try for the use of the network by that carrier’s become critical. This is important for in-building to propagate that through 300 feet of walls. You’re subscriber. wireless infrastructure, which means the going to have to be very localized that way, which Middleprise gets even more important, because comes back to infrastructure. It may become second nature for the network there’s more infrastructure that may need to go in operator to have these micro-agreements for each a building. If you’re using fiber and Cat cables, the enterprise. Agreements where they can handle infrastructure has to be something you can those users that go on the system, without having You might have an antenna on a conference table, constantly add to, just like Wi-Fi. Wi-Fi networks to fund the building deployment and manage the but it provides multi-gigabit speeds at that little today are continuously being optimized. They’re system. spot. It all connects back to the same gateway adding more locations, more access points. core. That ecosystem ideally would allow them to get It’s no different for licensed bands, either. There’s incremental revenue that they wouldn’t be able to It’s critical that the solution can scale to do all no reason why you shouldn’t be able to continue get if they didn’t invest in the whole system. these things, and that’s why a software-defined to add services as you go, which means that the architecture is critical.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |86| About SOLiD SOLiD helps people stay connected and safe in a rapidly-changing world through a portfolio of RF Amplifier, RF Radio and Optical Transport solutions. SOLiD enables indoor and outdoor cellular and public-safety communications at some of the world’s best-known and most challenging venues including leading hospitals; professional and college sports venues; government, university and Fortune 500 corporate buildings and campuses; international airports and metropolitan subways; and other high-profile sites. For further information on SOLiD DAS, Backhaul and Fronthaul solutions, go to www.solid.com or call 888-409-9997. About Ken Sandfeld As Executive Vice President, Ken Sandfeld leads the overall sales and product strategy activities for SOLiD’s portfolio of network densification solutions. Ken possesses over 17 years of experience in the wireless infrastructure industry and is passionate about bringing innovative technologies to market. Prior to his current leadership role, Ken held management positions at MobileAccess, Remec, Spectrian and Zyfer. Today Ken is focused on bringing SOLiD’s leapfrog technologies out of incubation and into the market to solve some of the industry’s biggest problems. Those areas include high-efficiency amplifiers for indoor and outdoor small cell applications as well as low-cost DWDM tunable optical solutions for the Enterprise and Wireless Operator markets.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |87| in the macro network dropped by 45%, freeing SpiderCloud is working to support LTE in the 5 GHz Profile macro resources for subscribers outside the unlicensed band, through LTE-U, LAA and venue. MulteFire. Unlike LTE-U and LAA, MulteFire does SpiderCloud Wireless not require a licensed band to anchor Services Nodes configure and optimize Radio transmission, so it will strengthen the neutral-host SpiderCloud has been a forerunner in the Nodes with SON, manage mobile access, and model in venues where multi-operator support is enterprise small-cell indoor market with its monitor QoE and performance. They are linked to required. SpiderCloud also supports the use of enterprise RAN, or E-RAN, solution. Initially E-RAN Radio Nodes via Ethernet connections that the Authorized Shared Spectrum (ASA), which fits well used 3G spectrum; now it supports any operator can install in the building or that are neutral-host and private LTE deployments. combination of 3G and 4G, as well as carrier shared with the enterprise LAN. In large venues, aggregation. SpiderCloud is currently developing the Services Node is typically installed on the Beyond the walls of the enterprise, SpiderCloud new products based on unlicensed LTE and 5G. premises. In smaller venues, it can be remotely offers SpiderNet, a centralized configuration, fault installed in the operator’s data center in a C-RAN and performance management system that SpiderCloud E-RAN uses macro spectrum (in a co- topology. enables mobile operators to manage multiple channel deployment) or a separate band to E-RAN locations. SpiderNet is based on Broadband provide better indoor coverage and higher The E-RAN topology accommodates MEC Forum’s TR-069, and uses IPSec tunnels to connect capacity for voice, video and other data traffic. The functionality, to support enterprise applications to the Services Nodes at the customer premises. E-RAN system has two elements: 3G and/or 4G that benefit from local caching and traffic Northbound, SpiderNet is integrated with the Radio Nodes (i.e., small cells) installed throughout management. operator’s core support system through NBIs. the building; and a Services Node, a small-cell controller that manages up to 100 Radio Nodes in a building up to 1.5 million sq ft in size and links to the core network.

According to SpiderCloud, a system of 100 Radio Nodes can support 100,000’s of data sessions and handoffs involving thousands of users; 32 concurrent 3G calls; and 64 active LTE connections at 150 Mbps. In early deployments with Verizon in the US, SpiderCloud reported that, after turning on an E- RAN network, the average number of connected users

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |88| My background is in the enterprise space. I got to first targets, but then as you go down in size, it’s SpiderCloud Wireless learn more about the wild world of cellular over like a pyramid. You have many more buildings, but the last three years. I do the marketing, they are smaller and, until recently, more difficult Making densification evangelization, a lot of writing and a lot of the to cover from a technology and financial point of public-facing activities for SpiderCloud, and I get to view. interact a lot with both our customers and our in the enterprise engineering organization, and the analyst Now we have the technology. How can small cells affordable community and the press. address the smaller-venue market, specifically? A conversation with Monica: It may be worth giving a bit of historical Art: With small cells, you’ve got essentially introduction, because enterprise plays a major role standard 3GPP cellular technology, but you’re Art King, Director of Enterprise in the client base of any mobile operator, and yet talking, let’s say, a 10,000 sq ft footprint. In the Services and Technologies, that segment has been somewhat neglected in case of Verizon’s deployment that we are terms of indoor coverage. How do you see that implementing, we are reusing the 30 MHz in the SpiderCloud Wireless evolving? LTE band 4/13.

Art: The enterprise hasn’t been neglected, per se. By reusing 30 MHz over and over again in the Monica Paolini: In our conversation with There was a lack of cost-effective technology to same building, you get a lot of spectral reuse and SpiderCloud, I talked to Art King, Director of address buildings of maybe under about half a extremely high capacity, because instead of having Enterprise Services and Technologies, about the million sq ft. that spread out over X sq m, it sits over only growth in indoor cellular coverage in the 10,000 sq ft. It changes the spectrum needs and enterprise. Within the higher end, there are quite a few allows operators to do a lot of the things they have buildings with DAS infrastructure, but a lot of been struggling with, by making the cell radius Art, can you tell us what SpiderCloud Wireless is buildings are too small for the fixed-costs to work extremely small. doing to improve coverage and capacity in the that DAS entails from a business perspective, for enterprise, and what you’re doing personally in both the operator and the customer. Small cells Monica: In the case of Verizon, you increase the this context? are opening up the total addressable market to efficiency of the spectrum asset by spectrum include smaller office buildings that just weren’t reuse. How do you achieve that? Art King: SpiderCloud Wireless builds indoor available in the past for coverage and capacity cellular systems. We call it the Enterprise RAN. It’s improvements. Art: We leverage standard cellular technology. a small-cell technology with Ethernet in the Cells were designed to interfere and have cell fronthaul within the office building, and Radio In the DAS model, you build a big wideband edges. It’s just the nature of the technology. We’ve Nodes distributed through the building. Its antenna through a building and then plug base built a lot of intellectual property around our footprint is very similar to the one you’d see in a stations into it. That model does not work for platform to enable a better experience for the contemporary Wi-Fi network. smaller buildings. We’re bringing technology to device owners. address the unmet needs of the subscriber We have the Services Node in the telecom heart of buildings from 50,000 to half a million sq ft. At the end of the day, the acid test from the the building. It coordinates the cloud of radios and operator community is really meeting KPIs. I don’t then ties back to the mobile core via high-capacity Monica: That’s a huge market. We always hear have a PhD in engineering, so the nuance of how IP backhaul. about stadiums and big campuses, which are the it’s really done is beyond me, but the KPIs and the

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |89| satisfaction of both the enterprise customers and the operators of the technology are the core message. We can tell through customer satisfaction that we have solved a lot of these problems.

A huge amount of automation has been added in the SON, the self-organization, self-optimizing software in our Services Node that coordinates the cloud of radios. With SON, the network keeps itself tuned and aligned and in harmony with the macro network in the outside world.

Monica: By making it possible for the subscribers within the enterprise to use the local network, you also offload the macro network. That means that the macro resources are free to be used in the wide area. How does the combination of indoor access and outdoor offload bring value to operators?

Art: Indoor and outdoor coverage were always looked at as two parallel things. That’s been one of the discoveries – perhaps not a discovery so much, but it’s been an internalized change in thinking in offloading them to a small-cell infrastructure and Art: We’re seeing both. some of the RF engineering groups – that indoor getting them off the outdoor network?” and outdoor infrastructure nodes complement We’re seeing operators that are saying, “We will each other. You’re seeing the emergence of densification not certify and allow SpiderCloud to be connected to only as a capacity and performance move, but also our network. Any reseller or enterprise that wants When we started seeing indoor systems drive a as a way to manage your spend on the outdoor to build out a network – as long as it’s done to our drop in macro usage by 50%, light bulbs started network at the same time. specifications and by one of our system integrators going on, where people started thinking about that will do a sanctioned, high-KPI installation – go doing holistic engineering. Monica: The change in performance with the for it.” offload has an impact on the business case. To When you look at a macro upgrade, you start date, the business models for indoor coverage There are other operators that are saying, “We looking at the dollars and saying, “That enterprise have been difficult, because it’s unclear who’s want to build and manage all the pieces.” That building right there generates a hotspot in the going to pay for the infrastructure – the enterprise obviously has its own capital impact and business macro. There’s a thousand people there. What’s or the operator. process impact. the cost of doing the macro upgrade versus

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |90| One big issue with releasing the reins of control is installation or surprise installation might cause channel interference with the macro. The bigger picture that is emerging is that as operators have fewer concerns about co-channel interference and more experience in managing it, they start becoming more comfortable with saying, “Yeah, let’s have the installations happen more freely, because we know that the technology doesn’t interfere with the outdoor network.”

RF engineers naturally start conservatively, because they are held accountable for the KPIs and the performance of the network within their geographies. I understand their concerns, but over time, things start relaxing to where you have the ability to sell to the enterprise and fulfill its requirements. There’s less of a concern about interference on the outdoor network.

Monica: Now there’s a lot of talk of neutral host small cells to light buildings where the Art: Licensed bands are running out of steam. You arrangements. Who do you see as the emerging performance isn’t satisfying our business users.” don’t have enough spectrum to keep up with the entities in the middle that facilitate these neutral downlink utilization by the customers. hosts models? You’re seeing both people talking about host neutral, in the context of MulteFire in the future, You added LAA to boost that downlink capacity in Art: You’ve got the large entities like ExteNet, but also enterprises that are saying, “I know I can’t the unlicensed spectrum. Beyond that, with Crown Castle, American Tower, and a lot of the get that now, but if I converge on to one operator MulteFire, you’re operating both uplink and traditional tower companies that are buzzing to do and manage my contract with them, I can satisfy downlink inside the unlicensed spectrum in a small cells. I have seen host-neutral Ethernets my business people.” friendly manner, without stopping Wi-Fi. Or if being proposed, where you can plug in small cells you’re operating it in the 3.5 GHz spectrum, you and, because they’re being installed above the Monica: You mentioned MulteFire. That’s an avoid the issue completely. ceiling, there are no aesthetic issues with multiple interesting development in the densification radios. universe. There is LTE-U, unlicensed, LAA, LWA, It’s a great way to solve the neutral-host problem. MulteFire. They all use the unlicensed 5 GHz band, Instead of building the wideband antenna and We’ve also seen enterprises that are basically and that tells you is that the unlicensed is of great doing base stations, you converge on the same saying, “We need cellular service indoors now. interest to mobile operators, especially in indoor piece of spectrum and you converge on the same What we’re going to do is to converge on one environments. What’s your view on these access methods of the spectrum. mobile operator. As part of our contract with that solutions? operator, we are going to request an allowance for MulteFire is being engineered to live in a contention-based environment, where you can

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |91| have multiple MulteFire instances running in the Subscribers don’t care, but as a vendor, an applications and a lot of content is tied to the same piece of spectrum, but you can also have a enterprise or an operator, you just have to find venue location. Is MEC going to be relevant to the MulteFire single infrastructure that connects to new ways to meet that demand. enterprise? multiple mobile cores. As we get up to 5G, and even before 5G, there is a Art: Yes. It could potentially be relevant in the Why is 5 GHz so interesting? It’s a planetary band. lot of talk of using different bands, all the way to future, as enterprises buy their own RAN Imagine if every UE in the world had a 5 GHz radio 60 GHz to 70 GHz, or 80 GHz. What is the scope for equipment. Right now MEC is going to be and MulteFire and it was available around the those bands with much more limited coverage for operator-controlled infrastructure, because the planet. It would solve the indoor problem for the indoor coverage? RAN is controlled by the operator. When MEC world in one equation. That’s obviously recklessly becomes relevant to enterprises – when they can optimistic and forward looking, but you can see Art: What we’re hearing on the product buy a box from a Cisco or an HP, and bolt it into a the end game being a great simplification of indoor management side of our organization, from the rack and connect – it could start being very infrastructure. CTO community, is that the operator community is interesting from the enterprise perspective. We’re looking at sub-6 GHz being indoor, and the very seeing plenty of traction and business interest Monica: Having the same infrastructure shared by high-millimeter waves are going to be outdoor right now with MEC. operators and fully integrated with their network solutions. They believe that what’s going on makes it easier to manage for operators. It allows indoors right now with under-6 GHz, because of In the context of what we’re doing, MEC is a more efficient use of the spectrum at the access just the sheer cost of building wireless access inserted in the S1 link between the RAN level, but at the same time, it may look similar to points at a microwave frequency, and having to equipment and the core. That can be either Wi-Fi. put a huge amount of them into cabling, it blows through a virtual machine, where we pass the up the business case. traffic up to the VM and then back to the core, or Art: No, it’s not competitive. In the enterprise it is in an external box. We’ve seen MEC being used complementary – envision the yin-and-yang circle. Sub-6 GHz, because it can generally propagate both ways. You’ve got Wi-Fi and you’ve got cellular. through walls and has decent indoor characteristics, will continue to be the direction, We’ve done quite a bit of prototyping with various One of our enterprise advisers said very clearly, “I with the very high frequencies being more of an software vendors, but the people who are in the don’t want to be a Wi-Fi provider. I want to be a outdoor solution. lead are looking at the holistic operating wireless provider.” He was very focused on environment, the maintenance, the patching, satisfying both kinds of wireless for his business Monica: The 3.5 GHz band is going to be a major everything necessary to run the whole life cycle of people because they were pushing hard for full opportunity for indoor coverage. MEC. These folks have thought through what are service on all pieces of technology. the killer apps that will justify deconstructing an Art: Yeah, definitely. Without any co-channel application out of a cloud data center into the In environments like the medical one, where interference in the 3.5 GHz, when you hand in, you mobile edge. you’re wildly overloaded in your Wi-Fi hand in to a completely different frequency. infrastructure, they’re looking to unload Wi-Fi by You’re not worried about the outdoor MEC isn’t a data center at the edge of the cellular getting mobile devices off of Wi-Fi. interference. network. It is going to be a purpose-built environment that has business value being at the Monica: From the subscriber perspective, wireless Monica: A new development is Mobile Edge edge. It isn’t put there just because you had some access from the device is the rule today. The Computing or MEC, which allows you push core computing capacity sitting in an enterprise rack. question is what technology the subscriber uses. functionality to the edge. Many enterprise

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |92| Monica: What are the enterprise applications for data out of the hospital through the mobile core emergency services calls can get within the MEC? and back to the tablet, because of HIPAA distance requirement of the FCC. regulations. Art: We’ve seen caching, local breakout, and a Monica: What are you working on over the next number of applications around the unified Whether or not that’s a misreading of the five years or so at SpiderCloud? communications integration, so that a lot of the situation, they were very clear that they had to native infrastructure on a mobile device can be keep that data in-house. They felt it was Art: Right now, the big things in the pipeline are used with the enterprise infrastructure. imperative to do it. It was a hard requirement. the LTE-U trials and the things that are going on They were saying, “How do I do private LTE? How with Verizon. One of the funny things is that it drags the network do I do it on the operator’s LTE network, but break behind it. There are retailers in the bricks-and- that traffic out and send it to my data center inside There’s a huge amount of interest within the mortar space that want to go to a wireless point- my building?” operator community that’s in our installed base, of-sale on tablets and mobile devices within their because they haven’t built any Wi-Fi stores. They’re seeking clean spectrum to operate Monica: Actually, that reminds me of another infrastructure, and there was a reluctance to build in instead of fighting in the mall with all the Wi-Fi issue about regulation: safety, security, emergency a whole parallel, separate mobile core SSIDs and all the adjacent shops and their wireless calls. How do you deal with that when you are in infrastructure to tie in Wi-Fi authentication. The APs. an indoor environment? emergence of LTE-U has generated a lot of interest in the operator community around the world for There are high-end retailers that are looking at Art: For our technology, the cell size and what us. this: “How can I move all these devices to LTE and we’re doing with our cells, each cell has its own get out of the traffic jam that I’m in right now, that ECGI. It’s very identifiable from an LTE perspective. Another thing we have going on is just executing can cause me to not be able to complete a sale the business plan, and scaling up and helping the and have to go to a wired cash register because When you look at average cell spacing, I would say operators we have in place get to higher volume, the iPad stops working when I am trying to take a a hundred ft between cells, we’re very much able and helping the system integrator channels get customer order?” to assist an operator in meeting the FCC more and more efficient and better at requirement that a call must be located within deployment. A lot of basic business blocking and Monica: At the same time, all those 50 m of the person that generated the emergency tackling to make the machine go. communications are local, in the sense that they call. are within the premises, right? The ECGI and the information that the installer Art: Yeah. I was in a situation where a hospital was sets when the radio is installed – this identifies saying that from a HIPAA perspective, they want to what floor, what zone on the floor – and provides use the LTE, but they can’t loop that customer what the FCC calls a dispatchable location, so that

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |93| About SpiderCloud Wireless SpiderCloud Wireless develops breakthrough, small cell network platforms that allow mobile operators to deliver unprecedented cellular coverage, capacity and smart applications to enterprises. SpiderCloud Wireless is based in Milpitas, California and is backed by investors Charles River Ventures, Matrix Partners, Opus Capital, Shasta Ventures and QUALCOMM. For more information, visit www.spidercloud.com and follow SpiderCloud on Twitter http://twitter.com/spidercloud_inc. About Art King As the Director of Enterprise Services & Technologies at SpiderCloud Wireless, Art leads the development of enterprise services definitions and business case propositions for customers and partners. Art is a Small Cell Forum Board member and a Vice Chair of the Services Working Group. Art was formerly the Mobility/Collaboration lead in Global Architecture for Nike Inc. where he held various global roles over 10 years. Prior to Nike, he led the build out of two multinational engineering and consulting organizations for an IP Services network vendor in the service provider industry. Art holds a BS in Computer Engineering from Portland State University.

III. Operators’ interviews

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |95| formed by the merger of Orange and T-Mobile, holes as well. Once you’ve got 4G coverage onto BT and both networks operated at 1800 MHz. We the macro-cell network, and you start to add took the two 1800 MHz networks, picked the best additional carriers, you can do the same kind of The many roads to sites in the portfolio, and had a fairly dense grid to thing on the smaller micro cells. start with in that regard. This will further increase an area’s capacity densification Of course, then as we added more capacity density. Generally, the location of these sites is A conversation with Andy Sutton, through 3G and on to 4G, we started to where demand is, because they were deployed for significantly increase the capacity density in the that demand during the days of 2G and/or 3G. As a Principal Network Architect, BT network. Our big focus to date has been around result of that, demand tends to grow in the same rolling out 4G to the macro cellular network. That’s kinds of areas, in the main. So, really good Monica Paolini: In this conversation with Andy involved upgrades to add 4G in a number of positioning in that regard. Sutton, Principal Network Architect at BT, we frequency bands. discuss the role that densification has for mobile Monica: What’s a micro cell compared to a small operators – and specifically BT in the UK, which LTE 1800 is our primary 4G band. In addition, cell? has recently become a mobile operator again after where we need capacity, we’re adding LTE 2600. the acquisition of EE, which in turn was a joint We’ve already launched carrier aggregation with Andy: Typically, in our network a micro cell uses venture of Orange and T-Mobile. two carriers. It’s commercially available on the macro-cell-type equipment. It’s generally got a network. We have a third carrier available to roll small cabinet at the bottom of a lamp post, or it is Andy, what is your new role at BT? out this is a second 2600 MHz carrier, therefore inside a building with the antennas on the outside, enabling three-component CA. but, generally, the microcell has a much lower Andy Sutton: Since BT’s recent acquisition of EE in output power than the equipment is capable of. So the UK, I’ve formally transferred into BT’s Chief Additionally, we’re rolling out 800 MHz where we it’s covering a very small area. It could be a smaller Architects Office, where I’m responsible for need it. One, it helps to enhance in-building version of that macro-cell base station in certain aspects of RAN architecture evolution, along with coverage. Two, it helps to extend geographical cases. mobile backhaul architecture and strategy. So I am coverage as well. Geographical coverage of the UK very heavily focused on the evolution of the is incredibly important to us, along with increasing I consider a small cell to be a freestanding, small existing LTE Advanced network towards LTE capacity. unit that can be mounted in its entirety on a lamp Advanced Pro, and working on our future plans for post that likely contains both the antenna system 5G. Once we’ve rolled out onto our existing macro-cell and the radio, with extremely low power – 1 to 5 network, then within that macro-cell network watts, probably in most cases towards the lower Monica: Can you give us an update on we’ve got a lot of what you’d effectively call micro end of that scale – and that would cover a small densification at BT? At EE, you were at the cells. These tend to be fairly short poles with area by design. forefront of the densification efforts for a long antennas on them, typically 5 to 10 meters high, or time. What has changed recently? installations on the side of shops, office blocks, Today, these systems tend to be a single etc., where we’re really down in the clutter. operator’s, quite often a single RF carrier. We need Andy: From the EE perspective, we had a very the ability to scale these solutions, certainly to dense network from the days of GSM, in as much A lot of these sites were deployed back in GSM support carrier aggregation, to match the as we were 1800 MHz–only operators. EE was days and evolved through 3G to add capacity. In capability we’re putting into the macro-cell layer some places we used them to fill in small coverage now.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |96| What we don’t want to do, of course, is hand users micro cell. It depends on the use case in the into a small-cell layer for capacity benefits, and deployment scenario. find that we’re restricting the service and the performance, which will impact the overall Monica: How do you move from micro cells to average network performance of the wide area small cells? network. Andy: The next phase is to really understand how What we want to do is invest in small cells that we deploy ever smaller cells, and tie that in, also, allow us to increase the average experience, be with our approach to in-building coverage. If a that data rate, reliability, low latency, etc. A small- huge amount of data is generated in-building, and cell layer has to be an answer to our problems and we can manage that capacity within the building, not something that’s going to restrict our ability to then of course it’s going to reduce the demand on enhance the quality of the experience. the external macro network.

Monica: Do micro cells have a single sector? Currently we’re considering the optimal strategy for balancing growth in external network capacity Andy: Not necessarily. They can be one, two, or with managing in-building coverage to remove, three cell sectors, depending on the configuration. effectively, capacity demand at the source, If they are on a lamp pole, for example, they could therefore helping us to balance the overall well have three cell sectors. In most cases, it would network. That also allows us to work towards a generally just be a single cell sector’s worth of more energy-efficient networks as well. equipment – not in every case, but in many cases. In areas of high in-building demand, rather than Therefore, they may go through a splitter to transmitting everything from outside in, we can multiple antennae for covering, maybe, multiple manage that capacity more efficiently with in- directions from a particular building. Or it could be building solutions. fully sectorized, or it could be a single stack of carriers sharing antennas. Monica: What challenges do you face as you move indoors? Monica: And do micro cells share the spectrum with macro cells? Andy: Among the challenges we foresee, site acquisition is always one that operators have. We Andy: Yes, all the same bands, effectively, planned need to develop new strategic partnerships. We pull now from the general public, from enterprise as part of the macro-cell network in that regard. also need to understand what new business and businesses, from academia, etc., to actually We’re not doing a lot to differentiate between the models are available for both external and internal have good-quality coverage and also capacity. layers. We do use certain parameters, such as deployments. hysteresis, for example, where you have to dwell New business models should make it simpler to on the micro cell for a certain period of time I think we’re definitely moving into a period access sites and infrastructure. In addition, of before you hand in to it. Again, that’s not on every where, rather than mobile operators pushing mobile communications on people, there’s a real

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |97| course, we need to make sure we can provide Monica: How will backhaul and self-backhaul backhaul connectivity to these sites. evolve as we move to 5G?

Backhaul is probably the second-biggest change Andy: In a 5G timeline, we need to understand after initial site acquisition. That’s going to require what we could do if we took all the various a combination of high-capacity, fiber-based spectrum bands available to us, all the way from solutions. It’s going to get expensive if we try and sub-6 GHz up to and including the higher run point-to-point fiber everywhere. We’re millimeter-wave bands that have been discussed currently exploring opportunities for the use of as potential radio interface technologies in the PON infrastructure. It could be an evolution of future. We need to understand how we could GPON. drop a small cell into a particular area in the network, and then have it develop its own Something I’m quite keen on is the introduction of backhaul using SON techniques. WDM PON in the future. With WDM PON, we could have dedicated wavelengths, so we’re not If we’re going to be using full-dimension MIMO or sharing the access between different use cases. 3D MIMO on these small cells, then potentially we could provide connectivity, either primary or Monica: How reliable – and valuable – is wireless backup connectivity, via a particular beam backhaul? between adjacent small-cell base stations, as well.

Andy: We have a range of wireless solutions It may well be that we have a primary link that is a available today for small-cell backhaul, from point-to-point 60 GHz radio, for example, or a traditional bands such as 28GHz to wider channels multipoint-to-multipoint system in a lower in the millimeter wave bands, including V-band frequency band. But, actually, if that link were to centered around 60GHz. fail or it needed more capacity, we could have alternative connectivity in the network. E band equipment is starting to get smaller. I think there’s further opportunity there. It’s still a little Monica: How is densification changing the way big and a little expensive at the moment for the you plan and deploy your networks? small cell use case. Maybe not so bad as a first hop from macro down, but in terms of distribution, it’s Andy: We need to think very differently about not quite there yet. how we build networks, because, as well as ultra- dense networks, we’re under increasing pressure As we look further forward into the evolution of to build ever more reliable networks – ultra- LTE and the introduction of 5G, we also need to reliable networks – and, of course, ultra-low- understand what self-backhauling can look like. latency networks as well. Today when we talk about self-backhauling, people think about near-non-line-of-sight or non- When considering densification, it’s important to line-of-sight and LTE frequency bands. But it does overlay a number of other requirements. Do a not have to be. densification plan. Do an ultra-reliable plan. Do an

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |98| ultra-low-latency plan. Then start to play those off We could couple them very tightly with something Monica: In the US, DAS seems to be much more against each other and understand what a single like LTE-U or LAA more specifically. There are a popular than anywhere else in the world. What do network looks like to meet those varying demands. whole range of other mechanisms we could you think about DAS? potentially use to help the two radio technologies From a capacity perspective, we’re either adding cooperate, as well. There are lots of aspects we Andy: We certainly deploy DAS in large more sites or more spectrum, or we’re getting need to consider when developing the overall deployments, large stadiums. There are better spectrum efficiency. We gained a lot of framework of a dense network. opportunities in large shopping malls. DAS lends spectrum efficiency by refarming spectrum from itself to shared deployments, where multiple GSM to LTE. Certainly more spectrum’s going to If you were just to build an ultra-dense network, operators want to invest, or where a site provider help, and more sites are going to help too. you would then have problems overlaying other wants to cooperate with multiple operators and requirements in the next decade on top of that. develop a single distribution network around the Then you end up with quite a complex You have to factor all that thinking into the stadium or shopping mall, for example. Those arrangement around acquisition and backhaul. The conversation quite early on. kinds of mechanisms work pretty well when the introduction of self-organizing network capability alternative is everybody’s trying to co-site all their is going to be absolutely key. The challenge I see Monica: You mentioned indoor coverage. Micro own small cells around the site. with SON today is that it tends to be focused on cells are mostly outdoors. How is it changing the aspects of the RAN or it can do something in the balance between indoor and outdoor? As you get into the smaller businesses, then you core, it’s not joined up at the network level. tend to be engineering more bespoke solutions Andy: Yes, indeed. We’ve got a number of based around that enterprise being a customer of Monica: Is network slicing going to make the RAN solutions for indoor. Starting from, obviously, large yours. In this case, the requirement from this more efficient? stadiums: England’s National Stadium at Wembley customer is simply to get coverage from your is actually connected by EE. We’ve got a huge network, in which case DAS is still one of the Andy: I’ve not seen anything that’s really lining up installation there: 24 cell sectors and somewhere options. behind this concept of slicing as we move through in the region of 200 antennas. LTE evolution to 5G. If we really want an end-to- When you get into the smaller businesses, small end service, we need to ensure that service is That’s an example of what we are doing at some cells may work micro p or pico cells could be available, that capacity is sufficient, that the low- high end locations, but actually, small cells for in- deployed and used with multiple antennas to latency specifications – whatever they happen to building includes a mix of pico cells and femtos. ensure sufficient coverage within the building, on be – are met, and that it is also reliable. Various other femto solutions are being used in a different floors etc. range of enterprise applications where maybe a If we want to go to multiple nines of reliability or large DAS system wouldn’t be appropriate. Monica: You mentioned that DAS lends itself to availability, we need to be looking at truly basically infrastructure sharing so you have heterogeneous networks with a whole range of As we’re addressing in-building coverage now in multiple operators on the same network. Could we technologies. This plays into the idea of using residential premises as well, we have launched not duplicate the model, making some allowances, for different radio access technologies as well, in both only VoLTE, but also VoWi-Fi. We have small cells? licensed and unlicensed spectrum, and Wi-Fi Calling now across a range of devices. If a understanding how to get licensed and unlicensed subscriber has Wi-Fi coverage at home, they can Andy: Yes. The small-cell story is an interesting bands to coexist and cooperate. make and receive calls on the Wi-Fi as well. one. When I mention sharing small cells I’m quite often thinking of the network-sharing agreements

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |99| that we’re starting to see in various places around on a per-operator basis against that radio operator to operate it, or is it the operator that the world. spectrum. Again, that ties into the siting, the pays for everything, as traditionally done? antenna selection, tilt, behavior, etc. They’re certainly well established in the UK where Andy: Traditionally it’s certainly been the operator you may want multiple operators deploying their Monica: In the UK you’ve had quite a long history that pays, but I think we’re all open to new models own carriers but on a shared piece of of having third parties working with you. now. It depends upon the enterprise’s aspiration. infrastructure. So having that flexibility is attractive. Andy: Indeed. There’s a range of operators. Fixed If the enterprise is a landlord wanting to guarantee and mobile operators themselves have a great good-quality coverage in the building to attract The alternative to that is for a third party to deploy opportunity to be a third-party neutral host. tenants, then clearly it’s in their best interest for small cells, and support multiple operators. That’s Developing some of those opportunities is them to be making sure that happens. The best a complex situation. How do you get the necessary something that, as an industry, we should be way to do that is to fund an installation, in part or agreements in place? How do you manage quality considering for sure. That would give the best entirely. of service, rollout priorities, backhaul prioritization, operational model. etc.? And then how do you actually do the hand- There are other approaches, of course, whereby if offs? We have a range of people who own large an enterprise is investing in Wi-Fi, the operator numbers of sites, for example, who are actually could deliver a solution that supports VoWi-Fi. These models have great potential, and we’re very buying up access to infrastructure. They could Then that could be part of the agreement – that keen to explore them. Some of the technologies potentially take on that roll or work in partnership the operator provides that service, makes sure all coming to market now are extremely innovative with mobile network operators or a wholesale the devices are capable, etc., and then the and offer flexibility to support a range of service provider. enterprise can make use of the Wi-Fi service within deployment use cases. the building for voice and data services. Conversations to date have always been a little I guess the question always is, does the operator challenging as to exactly where the demarcation is There is a range of different models we can want to hand over a level of control for the and where the operational responsibility is. As our consider here. It depends upon the use case and network to a third party, and how equipped are industry matures, everybody realizes that for the enterprises themselves. We get very different those third parties to take that role on, deliver that anybody to progress with these ideas and for ranges and requests. carrier-grade support that operators have been everybody to win, we have to find a consensus. doing for many years? We have to find a way of moving this forward. Monica: What about LTE unlicensed? How is that That means we need to be pragmatic in terms of going to play out in this kind of environment? Monica: In a neutral-host model, you could have a costs, contracts, and return on investments. situation where each operator has his own radio, Andy: In the UK, we wouldn’t necessarily expect so you would not have control over the physical Monica: With in-building small cells, is there a LTE-U, but certainly LAA is something of interest. infrastructure equipment, but you’d still control change in the relationship between mobile We’re very keen to understand how that will the spectrum and the access. operators and the enterprise? operate, how it will coexist properly with Wi-Fi. We’ve got to coexist within the established Wi-Fi Andy: Indeed, absolutely. I would expect that Specifically, traditionally the operator owns all the band. would be the case. We’d deploy our own spectrum infrastructure, but with most cells Wi-Fi type of an assets. The question then is just how many of the infrastructure, would you expect the enterprise to Certainly, having the ability to call up a 40 MHz parameters in the base station can be optimized pay for the equipment and then the mobile LAA channel in the traditional Wi-Fi band, the ISM

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |100| bands around 5 GHz, would give a huge boost in There’s a lot more spectrum coming in those At the moment, most of the expertise in these downlink capacity. So, using that as a bands. I guess the question is how big do you higher-frequency bands is in the fixed-link world, supplementary downlink and at 5 GHz, that really make a site, compared with how many sites do so it’s the people working on millimeter wave for lends itself to being used in-building, of course. you deploy? point-to-point, point-to-multipoint, multipoint-to- multipoint systems. Obviously, there are going to be restrictions on the You could, for example, build a large number of propagation characteristics of that particular band. small cells with an amount of spectrum on them, Integrating this into a radio base station with all They’re going to be difficult to match with other or you could continue to build out sites you have the associated multi-layer radio resource cellular bands. During trials, both in-building and today and expand them with more spectrum management in a heterogeneous network, and externally, it would be interesting to understand assets, more capability. also putting that into a UE with a decent antenna, exactly what can be achieved. Certainly that’s the is going to be a bit of a challenge. kind of thing that’s in our strategic plan. There are some practical site design constraints that limit what you can build, how far you can Of course, for it to work as radio access, it’s going Monica: What do you think about LWA? develop those sites. There are, of course, to have to be very tightly coupled with the lower- regulations around EMF that would restrict the frequency bands as well. It’s very likely the lower- Andy: LWA could be an easier alternative in some total power you could put out of those sites. frequency bands will provide the control planes regards. What we’re really waiting on at the and an amount of user-plane capacity. Thus the moment is chipsets to support this in a range of What we really need to understand as part of the bulk of the user plane could be pushed into a end devices so we can start doing some testing, millimeter-wave discussion is what does a dense millimeter wave radio access layer. and really understand what the options are. network of sub-6 GHz radio base stations look like? If we have smart antennas, we’ve got 3D Monica: How about using a C-RAN type of We haven’t nailed down one favorite at the beamforming, for example. Then we’re going to approach to increase densification? How big a moment, and I suspect we’d probably end up with get a step increase in spectral efficiency and area challenge are the latency and capacity a range of different hybrids of licensed and capacity density. How far does that get us? requirements on the fronthaul? unlicensed that meet different use cases. Something we’re working on at the moment is to Andy: The challenge for C-RAN today, of course, is Monica: What about millimeter-wave bands? You understand that evolution of area capacity density. the capacity requirements on CPRI’s fronthaul. mentioned the millimeter wave for backhaul. When we do move into the millimeter-wave Backhaul for a 20 MHz 2x2 MIMO FDD carrier What about for access? bands, then we’re probably looking at another would be typically 150 to 180 Mbps. With level of densification to support sites that maybe fronthaul, you suddenly jump to approximately Andy: With the current and projected traffic have a cell radius of 100 to 250 m. 2.5 Gbps to do the same thing with CPRI. volume, we will need millimeter-wave radio interfaces at some point, certainly high-centimeter Ensuring a robust radio interface connection in the That starts to burn up a lot of fiber and a lot of wave or millimeter wave. The question is, when do higher bands is a challenge and may involve a expense. So how scalable is that for your entire we really need it? range of techniques. Such techniques would likely network? Well, given the fixed relationship involve dual connectivity and tight coupling or between fronthaul today with CPRI, and a, the There’s additional spectrum in the sub-6 GHz coordination between cells in the same and spectrum bandwidth that you use, and b, the bands that will be made available in the near different layers of the network. number of antennas you deploy, that really plays future, as well as a number of other bands that will against wideband radio channels with ever more be made available over the next, say, five years.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |101| carrier aggregation. It also plays against massive Monica: How important is synchronization in a Monica: Optimizing and basically extracting much MIMO as well. multi-layer network? What level of more from what you have already, rather than synchronization do you need for enhanced inter- build. First, extract as much as you can, and then if I think the problem is scaling CPRI as we know it cell interference coordination (eICIC), TDD or it’s not enough, build more. today. I am encouraged, nonetheless, by the eMBMS? amount of work that’s ongoing, looking at Traditionally, it’s like a brute force. You try to send alternative functional splits between the radio unit Andy: The phase synchronization requirement is as much data as you can. The old networks didn’t and digital unit. typically stated to be around plus or minus 1.5 µs. give you a lot of opportunity for choice. As an industry, we’ve made good progress in As soon as you remove that need to split, as you understanding that, and developing ways to Andy: Absolutely. That optimization, traditionally, would do in conventional C-RAN, and start to deliver this in a robust manner, and support that has been something operators have done end- move a little further up the radio stack, we can level of phase alignment which we provide to-end on the network. As a wider industry, we start to offer some alternatives, something around increasingly, in addition to the standard frequency need to be talking more about what the MAC layer – split MAC for example – or synchronization. top-to-bottom optimization looks like as well. alternatively, split just before PDCP. Of course, the higher up that radio stack you go, the less optimal We need to understand, as LTE evolves and 5G is What kind of applications are we developing, and the radio coordination will be. introduced, where those requirements are likely to what protocols do they run over? A perfect go. Again, if we make it much tighter, we’re going example is the fact that over 50% of the traffic on There are still a lot of benefits to be had. Maybe, to add extra cost complexity to networks. We the EE 4G network is video. the ability to change that split dynamically need to be very careful about some of the between, possibly, sometimes the MAC layer, decisions we make about the level of coordination Of course, when video was first introduced to the sometimes the PDCP layer will give you a lot of and time alignment we need to achieve, so we internet, it ran over UDP. Now, it runs over TCP. flexibility to respond to different radio don’t create a huge cost and operational overhead That subtle change from UDP to TCP has a massive environments and radio channel conditions. for operators. impact on the efficiency of network operations. We need to look at top-to-bottom protocol-stack Certainly, in some environments, pure CPRI may Monica: This is an additional dimension when you optimization, and understand how we’re going to work, but I think that’s going to be restricted to look at the tradeoff between cost and evolve that protocol stack, over time, to a protocol stadiums and other environments like that. I think performance. architecture that delivers the optimal user really we need an alternative functional split we experience for end customers, network operators, can carry over, an Ethernet-based backhaul that’s Andy: Absolutely. That feeds into the discussion and application developers. going to have some benefits in the radio network. around SON as well. Given the number of base Actually, it’s going to be more backhaul friendly. stations we’re going to have and the complexity of What I like to think about in this context is not just those base stations, we really do need to start bps/Hz, but the application bps/Hz, as a true The performance requirements are not going to be automating more and more of these processes. measure of network efficiency. In terms of the as stringent as they are in CPRI, and the capacity We need to be making use of machine learning, so applications’ efficiency and the experience you get requirements, most certainly not. we can start to make more intelligent decisions as a result of that, and then how that ties into the about optimizing the network in real time. cost of building, operating, and scaling networks.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |102| About BT BT is one of the world’s leading communications services companies, serving the needs of customers in the UK and across 180 countries worldwide. Our main activities are the provision of fixed-line services, broadband, mobile and TV products and services as well as networked IT services. In the UK we are a leading communications services provider, selling products and services to consumers, small and medium sized enterprises and the public sector. We also sell wholesale products and services to communications providers in the UK and around the world. Globally, we supply managed networked IT services to multinational corporations, domestic businesses and national and local government organizations.

About Andy Sutton Professor Andy Sutton is a telecommunications network architect and designer with 30 years of industry experience. At BT, he is currently responsible for RAN architecture evolution and mobile backhaul architecture and strategy. Andy became part of BT after the acquisition of EE, having worked for EE since the merger of Orange UK and T-Mobile UK in 2010. Andy previously worked for Orange having returned to the company in March 2007 to take up the role of Principal Transport Network Design Consultant, he spent the previous two years working for 3UK, initially as a WAN Specialist and then as Lead Network Architect. Prior to 3, Andy was with Orange for 12 years and prior to that, he worked for Mercury Communications on fixed network transmission, switching and synchronisation systems. Andy is a Chartered Engineer (CEng), Fellow of the Institution of Engineering and Technology (FIET) and Fellow of the British Computer Society (FBCS), he contributes to the MEF (Metro Ethernet Forum) and NGMN (Next Generation Mobile Networks) Alliance on mobile backhaul topics. Andy is a Visiting Professor with the department of Computing, Science and Engineering at the University of Salford and a Research Mentor at the University of Surrey 5GIC.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |103| the concept of densifying our equipment. And frequency bands, the antenna spacing can be DOCOMO having many more cells coming closer together much tighter, and you can pack in lots of antennas changes the inter-cell-site dynamics. Remote radio into smaller footprints. Innovations heads, distributed MIMO, and of course massive MIMO are all very important. Monica: Right. Any difference between FDD and TDD for massive MIMO deployments? Technology enablers Monica: We have had MIMO for a long time, but it’s still evolving, with massive MIMO and Ozgun: This is a great question. It is very related to for densification distributed MIMO. How do you see MIMO the channel state information that we discussed. A conversation with evolving within the densification context? In FDD, the base stations send DL pilots to train the Ozgun Bursalioglu, Senior Ozgun: Companies and academic researchers see antennas for DL transmission. These operations Researcher, DOCOMO massive MIMO as one of the key technologies to scale by the number of antennas in the traditional enable all of 5G’s promises. Many people in way of doing FDD training. Now, if you follow the Innovations research have recognized this potential since traditional approach, combining massive MIMO Thomas Marzetta’s seminal work at Bell Labs. with FDD is not practical. Monica Paolini: Densified networks are not just First let’s remember one thing about massive In massive MIMO deployments, TDD becomes networks with a higher concentration of access MIMO. With massive MIMO, basically we have lots more practical, because training for DL can be points. They force us to think of network planning of antennas compared to the number of streams done through UL pilots. The user sends the pilots, and management in a different way. They require we are simultaneously serving. That just basically and the base station can train all of its antennas at and benefit from different technologies and makes the beams we are assigning to different the same time. The training cost scales in terms of architectures than macro-only networks do. This is devices very sharp. the number of users you are serving. That is, of the topic of this conversation with Ozgun course, great. Bursalioglu, a Senior Researcher at DOCOMO That means the interference between them is less. ovations in Palo Alto. That’s great, of course, for densification, because We should also mention that, in the FDD case, we can hopefully mitigate the interference there has been a lot of great research on doing Ozgun, what do you see as the major trends problem. This all assumes that there is high-quality FDD training with massive MIMO, but it requires a among operators as they work towards CSI, or channel state information, available at the bit more effort. It requires the learning of second- densification? What’s changing? What needs to be transmitter. That is a very important and deep order characteristics of the channel, and it works done? topic for massive MIMO. much better for the macro layer. Ozgun Bursalioglu: Here at DOCOMO Innovations, Monica: It’s taking MIMO to the next stage, and so In short, although there are some other methods we have research activities for 5G and beyond, that should lead to a much more efficient user that can be considered, TDD with reciprocity- especially on the PHY layer. Our recent spectrum. based UL training seems to be the preferred choice publications mostly focus on massive MIMO, and right now for massive MIMO. This is the reciprocity mmWave-band technologies. As for major trends Ozgun: Yes. Also, it has been shown that massive of the physical uplink and downlink channels. in densification, first of all, we need to serve many MIMO is very efficient in terms of power. It is the more users all at the same time. You’re talking way to go in 5G. With massive MIMO we have to Of course, this reciprocity from your device to the about increasing multiplexing gains, and this brings put up all of these many antennas, and in higher- base station does not hold if the RF chain is

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |104| considered. We have to do some calibration at the for a long time, because it is very high-quality It is a split of the control plane and the user plane. base station to take care of this problem. Actually, hardware. It takes longer for them to get de- You make sure the control plane is in the macro we and many other people have worked on that, tuned. layer, so at least the device is always going to be and it is possible. You can do an RF calibration at connected to the nearest macro cell and the the base station using signaling between BS If you have cheaper, more cost-effective devices, control plane information is available. antennas without using any collaboration from the you need to calibrate them more often. The other UEs. These are all very important points. thing is that you’re working with many antennas, If there is an established connection with the so of course it takes more time to calibrate the millimeter-wave band, we can get throughput Monica: I guess it’s becoming feasible because devices. from there. If not, we can always go back to the operators are able to optimize the RAN in real macro layer. time, so they have the information and then they We are trying to understand what grade of can just use it. Is it making it feasible to deal with hardware we are going to use, the number of Monica: Right. That’s very important, because you the increased complexity of MIMO? antennas, and what is the best way to do it. Also, can combine the reliability of coverage from the there is even calibration between distributed macro with the increase in capacity from the Ozgun: You have to calibrate the RF chains of antennas. This is all ongoing work. millimeter-waveband link. different antennas. All of these antennas can talk to each other and listen to each other’s signal. We have done theoretical research where we have Ozgun: There is one more thing. Now we have all That way they kind of correct themselves. They modeled hardware imperfections and come up these small-cell base stations, remote-radio-heads, align themselves that way. with algorithms to compensate these. There are and distributed antenna systems. In this case, the also ongoing trials, field experiments done by user, in theory, does not have to know where the There are some important differences, though. RF companies and universities. signal comes from. It might just need a unique cell calibration happens at a much slower rate than the ID that enables the device to connect to any training for the channel that we require. In that Monica: There’s a lot of excitement about MIMO, transmission point covering that location. sense, you can do the RF calibration of this but another topic that’s getting a lot of attention alignment between the antennas at the base now is millimeter-wave bands. What’s the best Monica: How does the unique cell ID work with station, and then you can go ahead and train the way to use that spectrum? small cells? What are the advantages of this channels and serve the users. approach? Ozgun: There are many ways. People are looking Monica: What part of this needs to wait for 5G, at using it for different aspects, but one way we Ozgun: It helps with the handoff procedures. and what part can be implemented ahead of 5G? look at it at DOCOMO Innovations is to increase When you densify the network, cells get very close throughput. It is great in that sense, just because to each other and the number of handoffs grows. Ozgun: Calibrating these antennas, of course, of the huge bandwidth available if you have a good Trying to switch from one base station to the next requires time and signaling design. I would say that connection. base station takes lots of time. It will have a huge we may need some standardization effort to overhead, and we don’t want that. reserve this signaling time for calibration. The problem is, getting full coverage with millimeter-wave is much more challenging, We would want a user device to seamlessly How much time we need depends on the tradeoffs because of the propagation conditions. We think it connect in a fluid way to all of the transmission that you are looking for. For the macro base is best to assist the millimeter-wave band layer points around it. Now, if you are trying to arrange stations, normally you can have very expensive with the macro layer. DOCOMO has suggested this the connection to a cell every time you realize, equipment where this calibration is not necessary approach, and called it the phantom-cell concept.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |105| “OK, I’m getting a signal from this cell,” it may and more important with real-time applications. already acquired the site for transmission, why just become a big mess if you do it very often. How can we lower the latency in 5G? put a single antenna? You could and actually should put many antennas to a site. Instead, with a unique cell ID, you have these joint Ozgun: There are many things that we want from points around you that are doing transmissions, 5G. Different applications may require lower Also, as you move to higher bands, to indoors, or but somehow you see them as a single point, like a throughput but very low delay; others may have a to hotspot scenarios, you don’t necessarily have single cell. That helps with saving the time, the high throughput but tolerate higher latency. these macro-style base station sites anymore that training, and of course, eventually the throughput. you can control. These new base stations can be The delay that we see at the user device, or in a anywhere. Monica: That’s an entirely new way to think about device without a user, such as a machine, is a cells as a unit. combination of things. The delay might be because In terms of capex and opex, you might need to of the application layer, the network layer, or the choose cost-effective devices. Although these Ozgun: Yes. Now, actually there is a trend towards PHY layer. devices are much cheaper than the corresponding user-centric approaches. We used to think about macro ones, you could still benefit from them base stations and cells, and the users in the cell, I prefer to talk about the PHY layer, and an using smart algorithms – for example, RF right? Then we had macro cell, cell-edge users and important part of it is coming from the traditional calibration. all kinds of other problems. way of training and associating to the base station – the way the base station sends the pilots, users Another example is that you might want the There is a trend, especially in the academic learn the channels, and the user sends feedback – distributed antenna systems to be synchronized in research, towards these user-centric approaches. all of these. terms of frequency or timing for coherent Users can be served by base stations (macro cells, transmission. You can do this with a common small cells, or DAS). And each user might have a With TDD UL training, which really sits well with clock, or maybe with over-the-air synchronization. unique set of these base stations that it can massive MIMO, training is immediate: the user connect to, so you don’t necessarily cut and piece sends the pilot, and the base station learns the We need to think about having networks where your network into cells now. channel. That saves you a couple milliseconds from these things can be adjusted. If the connection to the delay perspective. One important thing, again, one of them fails, your networks should be smart, But, again, this architecture needs to be scalable. at the PHY layer, is that as we are going to higher to use other available connections again. This architecture needs to be smart so you know bands, the coherence times get shorter. We need which base stations or which points of to be very careful about doing these things very It’s, again, very important to think not just about transmission to assign to this user, and this must fast. size but also the transmission points around the be very fast. We do research on these users. If one connection fails, your coverage should architectures. Monica: You can densify your network in many not fail. ways. You can use DAS, add small cells, or improve Monica: Increasing capacity is obviously the first the macro layer. What’s the balance there? Monica: Preventing failure is possible, because if goal, but there is also a need to reduce latency. you have multiple layers, the same mobile device Ozgun: Among all the things that you said, is covered by multiple elements in the network Because you can have all the capacity you want, basically the idea is really to have many more and has the flexibility to choose which elements to but if the latency is high, the service is not good devices and many more antennas. Even when we connect to. from a user perspective. Latency is becoming more talk about distributed MIMO, we can actually talk about distributed massive MIMO. If you have

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |106| Ozgun: Also, going back to massive MIMO, More users trained, more resources in terms of The base station I get my signal from should not because of its very sharp beams, a user device can time and frequency are given to the training phase just depend on the channel to that base station. It get data signals from many different points, unlike from the transmission phase. It is essential to do should also depend on whether that base station is the macro layer where you are connected to a these things in a more efficient way, in a way that overloaded with other users or not. single base station. you can simultaneously train users, instead of training each user one by one (one per pilot In the current system, the user device decides Monica: How are the requirements of dimension). We show that with pilot collision which base stations to connect to, based just on its densification changing or being driven by IoT? detection and fast user identification, you can channel gain. This system causes overloaded base overcome this problem. This way you can really stations, and that’s what we see in crowded areas. Ozgun: In IoT we might have different classes of make use of massive MIMO – because otherwise, If you can push some of the users, some of the operation. In one case a device might need to if you don’t do the training right, you’ll run into traffic, to the less-used cells, you can increase the connect to the network very fast. Maybe the pilot contamination or you’ll have to be really overall base station utilization in the network. connection can have a very low transmission rate, conservative and use a large pilot re-use, and that but the speed of getting connected is important. detracts from your densification gains. With massive MIMO, this is possible, because you Again, the delay is very important. have these beams that can come to you from Monica: In today’s networks, you have different different base stations. You should be able to pass The events for which these devices might want to layers, different networks, different access the stream from one BS to another one with less connect might be very random, very sporadic, or technologies, different bands, and so the question load. they might happen simultaneously. I think we is: how are you going to select which device selects need much faster random-access protocols, much which network? How do you decide all of these things while you faster training for introducing these devices into are transmitting in milliseconds? the networks they want to connect to. How can we go about it? The subscriber doesn’t want to manually pick a network, but from the A very important aspect of massive MIMO is that Monica: What are the crucial bits in 5G as far as network point of view, the operator needs to you have channel hardening. That means the densification is concerned? decide what’s the best way to serve that customer throughput you are going to get becomes sitting there with that device with some independent of the instantaneous channel. It Ozgun: If you think about it, we used to plan our applications that he’s trying to use. becomes independent from the small-scale fading. networks with our base stations, and decide which In a way, the channel hardens. users connect at which time, right? With the 5G, as Ozgun: You’re highlighting a very important part, I mentioned, connections to some of these which is load balancing, and, with that, user That means the base station can pre-guess, or the stations might be interrupted or there might be a scheduling. It is a very tough problem. In the operator can guess, the rate that the user will get blockage, so you need to be smart in arranging macro layer it is easier, in the sense that the cells from a base station. Having this information on a various connections. are bigger and you have a larger number of users. larger time scale than the transmission time scale Things average out nicely. is valuable. Since the traffic load changes happen At the same time, I think a major concern, much slower, you can optimize your network especially for hotspot areas, is the multiplexing Going to smaller cells or networks with multiple operations in terms of how to shift traffic between gain. We want to serve many users at the same layers, you’re actually seeing an irregular, maybe different layers. This optimization can even take time. That requires us to train all many of these an unplanned, network. Of course, the user into account the properties of the traffic – for UEs at the same time. doesn’t want to choose which ones it connects to. example, applications with different delay This needs to be automatic. tolerance. Channel hardening makes these

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |107| problems simpler, and we see lots of gains. Load you can have a pretty good idea of what you’ll get between millimeter-wave bands and the other balancing will be very important in the future. in terms of rate in the next milliseconds. bands.

For example, although the transmission happens That allows you to be smarter about where to put at the millisecond level, thanks to channel traffic, and it ties very well with the coordination hardening, ahead of time (on the order of seconds)

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |108| About DOCOMO Innovations DOCOMO Innovations is more than a just an organization for business development and strategic investment. All groups within DOCOMO Innovations work together to realize new applications and services through collaboration with American companies for the next generation of mobile services and beyond. DOCOMO Innovations provides vital innovation to enable the future growth of NTT DOCOMO, Inc. in the Japanese and global markets. DOCOMO Innovations is structured around four key teams, whose initiatives run the gamut from mobile applications all the way down to core networking technologies: Business Development and Investment, Open Services Innovation, Android Product Innovation, and Mobile Network Technology. About Ozgun Bursalioglu Ozgun Y. Bursalioglu is a Senior Researcher at DOCOMO Innovations Inc., working in the area of wireless communications on MIMO techniques and LTE enhancements since 2012. She graduated from the Ming Hsieh Department of Electrical Engineering, University of Southern California, in 2011. Her Ph.D. thesis is on joint source channel coding for multicast and multiple description coding scenarios using rateless codes. Previously she received M.S. and B.S. degrees from University of California, Riverside (2006) and Middle East Technical University (METU), Ankara, Turkey (2004), respectively. She received the best student paper award at the International Conference on Acoustics, Speech and Signal Processing (IEEE ICASSP), in 2006..

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |109| At the games, they want to be able to do Twitter, Monica: Before we talk about the technology, let’s Carolina Panthers Facebook, they want to take selfies. They want to talk about the ownership structure. Who owns do all the things they are used to doing with social and controls the wireless infrastructure? DAS and Wi-Fi join networking and connectivity in their regular lives. It got to the point where the number of fans trying James: That was one of the very first things I to get onto social networking and the internet in witnessed when I came here to the Panthers. We forces to keep sports general was exceeding the capacity of our systems. had a meeting with all of the carriers that were It didn’t matter whether they were using Wi-Fi or involved with the existing DAS system. We could fans happy the cellular systems for their data. They weren’t tell that there was dissatisfaction among those A conversation with James able to get through. carriers trying to figure out whether the problem was on their side, or with the DAS neutral host. Hammond, Director of That was leading to fan dissatisfaction. At about Information Technology, Carolina that point, they hired me – and one of the reasons We felt really powerless being in the middle. We was to tackle these problems. As soon as I got to couldn’t figure out exactly what was going on. We Panthers, and Kevin Schmonsees, the Carolina Panthers, I started examining the couldn’t figure out if we needed to put pressure on CTO, Beam Wireless systems that we had in place -- the DAS system for this or that carrier or on the neutral-host provider. rebroadcasting those cellular signals, as well as the With that lack of visibility, we just felt like we could Wi-Fi systems – and found that they both really never get to the core of the problem, and the Monica Paolini: Stadiums and other sports venues were not keeping up with fan demands. problems were never being solved. are typically the first targets of densification because of the extremely high usage density They were under-designed for a much smaller user In the meantime, all our fans were still during events. Our conversation about wireless population, making their five-year-old designs complaining that they were not getting the coverage at the Bank of America Stadium in obsolete. At that point we decided to bring on new coverage they wanted. We decided to move away Charlotte, North Carolina, is with James systems, and that meant ripping out and replacing from a system of managed service, which was Hammond, the Director of Information Technology the entire DAS and Wi-Fi infrastructure. essentially what our Wi-Fi and DAS were: we had with the Carolina Panthers, and Kevin third parties that owned and managed both the Schmonsees, the CTO of Beam Wireless. Monica: Kevin, what about you? Wi-Fi and the DAS systems.

James, stadiums are one of the best examples in Kevin: Michelle Rhodes and I started Beam We decided to let those contracts expire and then terms of high usage creating a highly challenging Wireless when we saw a gap in the industry where take over the ownership ourselves. We tried to environment. Can you give us an introduction there was frustration amongst venue owners just find the right consultants, such as Beam Wireless, about what you do personally and what like the Carolina Panthers. They just don’t have the to point us in the right direction, give us the advice infrastructure you have in the stadium? visibility and control of a DAS inside their venue. we needed so that we would have the expertise on board working with us, while we would James: At Bank of America Stadium, we were What we try to do is provide the consulting – maintain ownership of the system, and have full faced with a problem over the previous two or whether it is design, contracts with carriers, or visibility, and take full responsibility for it. three years where our fans were not able to get validation and optimization, to allow them to the connectivity that they expect. learn, manage, and have full visibility into a DAS Monica: Right now, are Wi-Fi and DAS integrated system in their venue. and you manage both of them?

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |110| James: In the case of the DAS system, we Once we made that decision, started planning, enjoying the experience of a sports event. How has contracted with Beam Wireless to help us select a starting building, we realized it was the best this experience changed through the years? DAS replacement and help us do the construction, decision to make. I think even the carriers agree integration, optimization, and ongoing with us that moving to this new model of DAS James: New social networking apps come out maintenance. DAS is pretty complicated. DAS is ownership was the right choice. Because we are every day, right? Each one of those leads to more complicated than Wi-Fi when it comes to the taking that full responsibility, we have high challenges. For example, a couple years ago, I had management and maintenance. expectations that carriers then can get the benefits not heard of, or perhaps it did not exist, Periscope of a good system to participate in. – or the ability to do live video streaming. New In the case of Wi-Fi, we also selected an integrator apps like that pose real challenges, because with called AmpThink. They came in and managed the Monica: How many carriers participate in the video streaming we’re getting into high bandwidth construction and integration, and they are DAS? usage. assisting us with our first year of service. Then we’re going to continue with the maintenance James: When we built the new DAS system, which Then there are also rules about what you can do in ourselves. was last off-season, in 2015, we maintained the a stadium. We’re not supposed to permit a video tenants that were there before on the old systems stream of a game going out live because of usage Although we own both systems, I think the level of – Verizon, AT&T and Sprint. We’re currently talking and broadcast rights. support we require from third parties to help us to T-Mobile about also joining the DAS. manage them varies between the two. There’s always something new coming out every Monica: In terms of the usage model, the wireless day. But more importantly, the volume and Monica: The decision to take ownership is difficult, connectivity has become really an integral part of density have changed. Before, you would say it because it’s a lot more work on your end. Is it was a small percentage of our fans that are getting because you think connectivity is something so crucial to you and for your fans that you want to have control?

James: It is a scary thing to bring a third-party managed system in house and take full responsibility for that. That was a scary decision, but it was the right decision to make. Because we’ve got a saying here. Our owner, Jerry Richardson, says that the fan is the most valuable member of our team.

That’s really important, because what it shows is that his emphasis is on the fan experience. If we’re finding that fans are not getting the connectivity they demand, then we need to fix it. And to fix it, we really felt we just had to take full control.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |111| on and texting and tweeting and taking selfies. But But subscribers are still takers, not givers of as our digital natives get older and older, we’re content. That’s kind of the expression that they getting more and more of them coming into the use. Definitely watching it. football stadiums. Monica: What is the balance between cellular and Overall it’s really the percentage of users wanting Wi-Fi? to get online that has changed the most. James: I was asked this question many times by Kevin: But also one of the big trends we see, and management here at the Panthers, trying to figure we see this not only in the Panthers stadium but out, “Hey, if we make our DAS so wonderful, why most arenas and venues across the country, is the do we even care about Wi-Fi?” Or, on the other trend from voice usage to data usage. hand, “Our Wi-Fi is really great. Why do we need DAS? Because I don’t need to get onto the cellular It used to be making a phone call; now it is sending for data.” a text message. That’s how fans communicate during these events. Now it’s Snapchat, Instagram, I do not see how you can avoid having both. In a Facebook. It’s more of the data usage and the large venue you’re going to have to have a good multimedia on the phone. You can see that in the DAS system and a good Wi-Fi system. DAS statistics. First of all, each system can only go so far. If you’ve One key thing the Panthers gained by having got two different systems, that right there is going control and management of their DAS is that they to help your density problem. get to see each operator’s statistics after every event. They can see what the true fan experience Second, you’re going to have expectations from was per operator after every event. For example, if the fans. They may come into a stadium with the lower bowl has a fan experience problem in a excellent DAS. They may realize, “Hey, I don’t want Kevin: I’ll say, “Please use it.” It’s funny: because I certain area, they can focus on that and try to to pay for my cellular data.” Not everybody’s got handle the DAS side, the expectation would improve that for the fans. They didn’t have that free cellular data, so they’re going to want that probably be that I wouldn’t care that much about visibility before. free Wi-Fi. I think you’re going to have the fan Wi-Fi. But I am thrilled to see the new Wi-Fi system demand driving the need for both. You’re going to going into the stadium, because the biggest help Monica: Some of the applications you mentioned have the overall density requirements driving the for the DAS is to have that Wi-Fi data offload. are also very heavy on the uplink. Do you see a need for both. change in the uplink versus downlink balance? It’s extremely important to have both systems in Monica: Are you looking into using LTE in place. We need to offload to Wi-Fi. Wi-Fi needs to Kevin: That’s something we’re definitely watching. unlicensed bands? LTE-U, LAA, MulteFire? offload to us. Not everybody’s going to attach to a That’s a great question. I haven’t seen a significant Wi-Fi system, and like James said, not everybody’s change yet. There’s still a big load on the downlink. James: I’m going to start by saying, “No, I’m going going to want to pay for data to stay on the We’re very curious. So far we’ve had two events to need the spectrum for Wi-Fi.” But I’m going to cellphone system. They really have to work this summer. I’m curious to see how the uplink let Kevin continue. together. It’s extremely important to have them ticks and does the downlink potentially go down? both.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |112| James: We built the new DAS system first in the Monica: How did you pick the DAS system you didn’t have this capability. It’s really come in handy off-season of 2015, and the new Wi-Fi system was have? now. installed during the off-season of 2016, so we just finished that construction. James: At about the time that I got in contact with There are also a lot of maintenance advantages Kevin, here, from Beam, we were able to quickly that we talked about on the whiteboard. But as we were doing that construction, I was narrow it down to three different DAS types. Then Something very important is the PIM test. Every approached by two of the carriers, wanting to talk I did some field visits – went out to other stadiums year, we go through an audit and make sure all the about what kind of agreements they can work out and took a look at what they had, and we passive infrastructure is still in good shape. with me in order to do the offload. narrowed it down to the final choice. For example, take the antennas, the cable and the These are the same carriers that are on our Kevin: We had a long, in depth whiteboard splitters. How have they been impacted excellent DAS system, so you can see they’re very session, and that’s what I encourage any venue to throughout the season? We can actually put this interested in all systems working together, do when it wants another system, is to have this PIM tone through every amplifier in the system because they recognize the density problem. You technical whiteboard session. James and I sat very quickly and efficiently – we can even do it have only so much space in Wi-Fi. You have only so down and we drew out every manufacturer and remotely and test all of that passive infrastructure. much space in DAS. They definitely can every product line for that manufacturer to discuss complement each other. what the pluses and minuses would be for the Monica: Did you choose a passive DAS? Carolina Panthers. Monica: When you decided to build your DAS, Kevin: No, it’s an active DAS. Our DAS has an active were there other options, such as small cells, that The product we selected might not be the best amplifier that’s fiber fed and then amplifies you considered? product line for another football stadium. But then through the antennae. another vendor’s product line might not be best Kevin: Macro won’t be able to cover the stadium. for the Carolina Panthers. Monica: No matter how good your system is, with There’s too many fans. One or two sectors coming Wi-Fi and DAS, at some point you’re going to see in from an outdoor macro is not going to be able One of the reasons we selected the product line congestion, because users continue to find new to handle the capacity. That’s not really an option. that we did was that it comes in what’s called a ways to use available capacity. What’s your NEMA-rated form factor. We needed to put experience with that? You can’t avoid congestion, “Small cell” is really a loosely used term. A small remote amplifiers outdoors, where they could be but you can manage it. cell originally had a very minimal capacity. Same exposed to rain and wind, cold and hot weather. thing: that’s not really feasible for a football Kevin: Right now, the DAS is fairly new, and we’ve stadium. We needed an amplifier that was a little bit more been through one whole season. The DAS is not robust as far as power, because all of the carriers going to be the bottleneck. It’s basically a very The newer term that people use for small cells is were going to share that power. We weren’t going wide interstate or highway that can handle any really a “lower-power radio.” That is actually to have separate amplifiers, because of cost and traffic that the base stations have been sending what’s used to feed the DAS. What a lot of people aesthetics. through it. call a small cell now is a 5W radio. But the hardware referred to by original term “small cell” Optimization features were also important. For The bottleneck right now is the operator base was used, say, for an office building or something example, we can attenuate every remote amplifier station. For example, the DAS as it sits today is like that. It’s really not feasible for a bowl of a per band during optimization. That’s extremely designed for 48 sectors of traffic per band, per football stadium. important, because a lot of the older product lines operator. The most we have running today is 32

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |113| sectors from one of the operators. Last season the halftime, which is the busiest time during an event, efficient anymore. That’s where additional most that we had was 24 sectors from one of the the device just doesn’t seem to work anymore. frequencies come into play. operators. What we see then is the payload doesn’t increase, James: The simplest thing we can do to address The bottleneck is the sector limitation on how but the attempts increase. What you see is people congestion is to ask the carriers to simply give us many users their radios can handle. For example, continuously hammering the system trying to get more. In other words, if we have 48 zones that are last year, if we take one operator, our failure point through. available to them right now in our infrastructure, was 15 GB on the LTE download in 15 minutes. they need to use all 48 rather than simulcasting Monica: If you look at it five years from now, their sectors across that. This year, that same operator has gone from 18 you’re going to see a much bigger need for sectors to 24 sectors. Now they’ve exceeded 28 GB capacity. How are you keeping up? The first thing is the easiest thing, and that is in the same 15-minute period without failure. simply tell the carriers, add more capacity. We can Without changing the DAS from last year to this Kevin: It’s a great question. Like I mentioned, take that. This great big highway can take it. As year, they just increased the base station capacity today the most that any carrier has is 32 sectors on Kevin said, you just need to add more lanes, and by increasing the number of sectors. the DAS. The DAS proper was designed for 48 that’s really on the carriers. sectors. When we talk about what it was designed Monica: When you have congestion, from the user for, what that means is SINR, signal to Monica: From your point of view, you’re ready for perspective, what do the people see? For how interference-plus-noise ratio, and channel quality more capacity as long as the operators come in long does the network go down? indicator, or CQI. They basically apply to the fan and make the investment. When they see the experience. When we look at statistics after an unique capacity, they’re likely to make that Kevin: We have to talk about congestion in two event, if the CQI in a zone is bad, the users in that investment, because it’s clearly to their advantage manners: voice calls and data. We don’t see any zone have had a bad experience. to provide it. voice congestion on the path at all. And I don’t think we have to discuss voice because the trend What we do is we design a system that’s, say, 48 Kevin: We personally discuss that with them. We really has gone from voice to data. sectors for a certain SINR level. That’s based on a have everyone’s statistics, and we can compare certain percentage load on the carriers’ base them. We can really push them in a direction, For LTE data, typically what happens is that our station. Once it exceeds that, then we’ll have to saying, “Hey, you have a capacity issue, in this uplink tanks; that’s the first thing we see in the come back and modify the DAS. concourse and seating area. Next season or mid- stats. What we see then is that uplink throughputs, season, you need to double your capacity in this which normally would be several mbps, suddenly In the bowl, we’ve pretty much hit the limit. The area.” Again, the DAS is prepared for that. go down to 0.1 or 0.05 mbps. only way to get more sectors in a bowl is adding new antennas, which vendors come out every year James: If we’re doing our jobs right, and the The noise rise in that zone is so high that either the with different beam widths, new bands that the Panthers are, what that really means is we’re customers can no longer communicate with the FCC licenses. Whether it’s the WCS band on AT&T, going to have little or no congestion. Rather than base stations or their throughput is so slow that the new AWS bands for T-Mobile, Verizon and accepting that congestion is going to happen, we the information never gets through. AT&T, or Sprint’s BRS band, all up in that want to have little to no congestion, because we 2.5-2.6 GHz range, that’s what will happen. want to meet the fans’ demands. To a customer, what happens is all of a sudden it’s Whether it’s RF or noise, it doesn’t matter. We can as though it doesn’t work. Their data worked 15 only put so much of it in one section before it’s not minutes and then all of a sudden, during, say,

About the Carolina Panthers The Carolina Panthers is a professional American football team that competes in the National Football League (NFL), as a member club of the league’s National Football Conference (NFC) South division. Announced as the NFL’s 29th franchise in 1993, the Panthers began play in 1995. The team is headquartered in Bank of America Stadium in uptown Charlotte, an outdoor 75,412-seat stadium that serves as the team’s home field. The Panthers are one of the few NFL teams to own the stadium they play in. The Carolina Panthers have had two Super Bowl appearances, won two NFC conference championships, six division championships, and seven playoff appearances. About James Hammond James Hammond is the Director of Information Technology for the Carolina Panthers, leading a department of nine staff members who support all technology aspects of the Panthers’ operation, as well as wireless, Wi-Fi, and network infrastructure for Bank of America Stadium in Charlotte, NC. As part of an initiative to significantly improve the Panther fans’ connected experience, he coordinated the selection, construction, and implementation of a new Distributed Antenna System (DAS) for the stadium, and is currently involved in construction of a new Wi-Fi system consisting of over 1,200 access points. He has also coordinated other major projects, including migration to a new phone system, and implementation of a fiber GPON solution that was part of the renovation of technology in all club suites. In 2013, Mr. Hammond was featured on CNN discussing the advantages and disadvantages of iris scanning and other biometric identification methods. Mr. Hammond previously served as the CIO at Winthrop University and has taught Computer Science at Winthrop University, University of Maryland European Division, the S.C. Governor’s School for Science and Mathematics, and Rutledge College. He has been admitted to candidacy for the Ph.D. in Computer Science at the University of South Carolina. About Kevin Schmonsees With over 17 years of RF design, performance, and optimization experience, Kevin brings a passion for improving the wireless experience in a DAS environment. Having worked for both wireless carriers and integrators, Kevin has extensive hands-on knowledge with 1st-class designs, carrier relationships, installation/implementation, integration, commissioning, and optimization. Kevin's goals with Beam Wireless, Inc. are to bring visibility and control to their clients- allowing them to focus on the overall needs of their customers through a seamless experience. He has a BS Electrical Engineering from NC State University and an MBA from Strayer University. Kevin can be reached at [email protected].

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |115| our internal security group. We need to make sure I have three different brands of antennas to supply Enterprise, all those come together. the three different carriers I have. That is a very big challenge for my facilities department. On the anonymous Monica: What wireless infrastructure do you technical side, having those three carriers use currently have? different systems means all I am getting from them is cellular coverage. A view on GM: We are a large Wi-Fi shop. We deployed that a long time ago. We have different, varying If I had one cohesive system, then I might be able densification from densities of Wi-Fi. Cellular has always been a to pick up some additional benefits, such as the challenge for us, in our buildings, especially as we ability to take calls off of that system and handle the enterprise move to LEED-certified buildings. the call control ourselves. I have three different systems. That’s never going to happen. A conversation with a Director of We opted for a DAS probably 10 years ago. That’s Network Architecture and served us pretty well, but we’re seeing limitations Monica: Do you need to have indoor coverage as far as density is concerned. We have a very from multiple operators? Wouldn’t one be Strategy at a US-based global large campus and we keep expanding it. Every enough? manufacturing and retail time we add a new building, the experience for the rest of the campus goes down. GM: One of the carriers has about 80% of our corporation traffic. Secondary carrier has about 17% of it. The As our populations inside these buildings are rest is a couple other small carriers. So, in the US Monica Paolini: This interview is with a Director of getting denser, we’re seeing that that, too, has a main campus, we need at least two carriers. Network Architecture and Strategy at a degrading effect on the wireless signals. multinational corporation headquartered in the Monica: Do you have DAS in other locations as US, that includes multiple business units in the US Monica: Are you planning to replace the DAS with well, or only on your main campus? and abroad, involved in designing, manufacturing, more cells, or to extend the DAS in your main marketing and retail. The interviewee has chosen campus? GM: We have DAS in multiple locations. Anywhere to remain anonymous and we will refer to him as there’s been a large problem, with enough people GM. GM: I’m looking to move to small cells. I’ve been complaining about it, we get a system in there. pushing carriers in that direction. Unfortunately, I GM, can you tell us what is your role within the haven’t got them on the same page as I am. They A lot of these places are smaller, so we don’t do a enterprise IT/communications organization? have their single-point solution; for each carrier, baseband headend. We end up bringing in the it’s all different. signal from outside, which we have varying luck GM: As a Director of Network Architecture and with. Depends on how bad the cell coverage is Strategy, I need to make sure the technology I’m OK with having different systems for different outside, and how overloaded the macro cell is. trends in the industry are going to match up with carriers, but I have a problem with the varying That’s been a mixed bag. the business needs for all the different businesses looks of those systems. That sounds weird, but I we have inside of our company. And that covers have a very particular facilities department that Monica: You said you have to work within an the network, unified communications, voice, and goes for a very specific look in each of the offices. environment in which you have multiple touches heavily on security. I work very tightly with operators, each coming with their own

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |116| infrastructure. What kind of alternatives are you bringing in the signal from the outside for the comfortable with.” This is very new for them in the considering? other carrier. Even though we have both carriers US, and so they’re looking at it and wondering, “Is on our DAS, still one has a better experience than this is a solution I’ve vetted? This is what I can GM: What I’d really want is one provider that the other. You go to a different building, that prove. I know this would work, but I don’t know if could give me one system on which I could put might be reversed. it works when I have all these other carriers on multiple carriers. I don’t necessarily want to run with me. I haven’t seen that.” the system myself, but I want the benefit of being Trying to tell VPs about that, they don’t quite get able to add services as small-cell companies add it, and they ask, “Why can’t we just fix it?” After small cells get out there and people demand features. them we’ll hear: “Oh, we can manage this. We can Monica: Do you have a neutral host for your DAS? maintain this.” The carrier is not going to offer them at the same rate I want to consume them at. Additionally, if GM: Yes, we do. Monica: One of the key discussion items is about each carrier used different solutions I would not be who’s going to pay for this infrastructure. Who do able to offer the same enterprise features for all Monica: As you move to small cells, is the access you think should be paying for the equipment and employees. going to improve? for installation?

Monica: You want to have consistency in your GM: Each carrier says, “We will not share with the GM: I feel that, depending on the level of control, I network. On the service part, you want to have all other carriers.” The reaction to DAS was very don’t mind owning it. I may not want to run it, I operators being able to support the same services. similar, in the beginning, to the one I’m seeing in may want to push that to a third party, but I do What is the challenge there for operators? small cells right now. When DAS first came out, want to have full control over what features are every carrier wanted to run their own system and turned on. GM: In the area I’m in, Verizon uses Ericsson. AT&T did not want to combine the systems. leads with a Nokia solution. Because they’re two Monica: Basically, what you’re saying is that “I’m disparate solutions, they offer different services. As DAS matured, we saw more and more carriers willing to pay as long as I have some level of allowing other carriers to be on their system, as control over what goes on, and what kind of The technology actually may offer some of the well as allowing a third party to come in and put in services the operator provides.” You put in some same services, but again, carriers only implement the system and have any carriers join. The carriers money, and you want to know how that’s being the ones they’re comfortable with and they’re are even pitching some money for that. used. What’s the response from operators? willing to support. That means that, even if the technology could support a service, I can’t leverage Right now with small cells, we’re in that very first GM: With the DAS system, it’s a shared model it, because the carrier decides not to offer it. phase, where all carriers want to put in their own right now. I pay for some of it, the carriers pay for system and they don’t want any integration with some of it, and I have no flexibility. I can’t see Monica: In addition to this bifurcation at the other systems or vendors. anything that goes on with that system. service level, you have a similar bifurcation in the access, because you have the different equipment Monica: Why don’t they want to share the small- Monica: As an enterprise, you may cover the cost from each operator. cell infrastructure? They are used to doing so with of deploying the network but, as you said, you may DAS. not want to operate it. Who do you think is best GM: Correct, and we’ve actually seen that in our positioned do that? Is it the operators, or is it a DAS as well. At some of our locations, we have a GM: I think it stems from the carrier mentality that third party? baseband headend for one carrier, and we’re “I do not want to support something I’m not

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |117| GM: I say it’s a third party. We have a company anybody interfering with the spectrum plan that Monica: The first step is coverage and capacity. right now that comes in, and they’re the ones that we put together. Would you say that it’s more coverage, or is it work with all carriers. They look at the costs and more capacity? negotiate deals with carriers. Carriers will only Monica: Your workforce is becoming increasingly pitch in if it means something to them. They put mobile. Your employees can be anywhere. You GM: Capacity. As we go to the more open-space, that case together. This company knows the could be working from home or a coffee shop off the more densified employee footprint inside the carriers and the licensed RF space. I don’t have campus, anywhere. How is that changing your buildings, and more mobile, we are getting rid of people that are in-depth with that; that’s why I wireless strategy? desk phones everywhere. I don’t think soft phones look to a third party for that. are going to pan out. People are walking a lot, and GM: On the Wi-Fi side, which doesn’t come into are getting rid of laptops. When you’re walking, Monica: Will the operator or the third party be the licensed carrier spectrum, I’m looking to create what’s the only device that really works? It’s your able to share your indoor wireline infrastructure the consumer experience. One of the largest cell phone. for backhaul or fronthaul, or do you expect them complaints of any enterprise customer is “It works to build a separate network? better at home.” What I want to do is give our Monica: Is latency much of an issue? employees consistency with how they connect to GM: This is where I’m going to differ from the Wi-Fi and LTE whether they are inside or outside GM: I tend to look a little bit further out. In director of security. I don’t mind having carriers on our walls. working with my application teams, they’re of the my infrastructure. I feel I can isolate them and same mindset as I am, that they want the keep the rest of the network secure. However, I do Monica: For you, the challenge is not when your experience to be the same outside as inside. need to get security’s buy-off on that. That’s going employees go outside the campus, it’s when they They’re building applications for that. to be one of the challenges we face. are on campus that it becomes the major issue. You look at the Google Play store, Apple’s store. I think we can come together. We all see the value There is a lot of talk of how operators can offer The apps that are being developed don’t care of it, and we just work through those issues. I’m services that are for the enterprise, and better where they are. They’re built for that user not sure how that’s going to play out yet, though. access. What do you think operators should do experience. No matter where you are, they’re differently to support the enterprise better? latency tolerant, and we are designing our user- Monica: What about carriers using LTE unlicensed, facing apps the same way. either LTE-U or LAA, in your campus? GM: First one is, solve my problems. My problems are that I need cellular coverage that’s a lot more Monica: What about Wi-Fi Calling? That might GM: That is a very big concern of mine. When we robust than we have now, a lot more throughput improve the voice quality, but then you’d have to first started looking at small cells, carriers wanted than I have right now, and seamless access inside allow the operators to use your Wi-Fi to pair it with Wi-Fi directly. They wanted to run and outside the walls. When I get to that point, we infrastructure. the Wi-Fi for me, or just layer on Wi-Fi on top of it can talk about what are the things we can leverage as a guest. Carriers are very good at licensed on top. GM: The issue I have with Wi-Fi Calling is that I spectrum. Unlicensed, I don’t think they have as would need a user to turn on a feature that will be much experience there. If they want to be a partner and come to me with affecting them even when they are not at work integrations to the rest of my infrastructure, solve and I cannot guarantee the experience. If they turn Frankly, my enterprise people know that my pain points first and then they can definitely on Wi-Fi dialing to achieve a good experience at unlicensed spectrum pretty well. I don’t want have a seat at the table. work but it creates a worse experience for them at home, we are doing a disservice to our employees

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |118| Monica: What do you think are going to be the one of my offices and go to a software as a service, GM: It’s more than multiple devices. If you have main changes over the next five years? you go directly there. I don’t need to transport you one device with 10 different apps on it, each one there. of those apps may be going somewhere different. I GM: My main changes are going to be any-device- don’t want to nail you down because you have one to-anywhere. Not force people to come through a Monica: You’re saying multiple devices, but also device and one person, to one point. I want you to central area or a central location to get to what multiple devices for the same person. It’s not more be able to go everywhere direct. they need to. More things in the cloud, more types but every person would have multiple things distributed, and if you need to take off from devices at the same time.

1080p 1080 pixels (or Full HD) DSn Digital signal n IEEE Institute of Electrical and 2G Second generation DWDM Dense wavelength-division Electronics Engineers 3G Third generation multiplexing IMS IP multimedia subsystem 3GPP 3rd Generation Partnership Project ECGI E-UTRAN Cell Global Identifier IoT Internet of things 4G Fourth generation eCSAT Enhanced Carrier Sensing Adaptive IP Internet Protocol 4K 4,000 [pixels] Transmission IPSec Internet Protocol security 5G Fifth generation eICIC Enhanced ICIC ISM Industrial, scientific and medical AP Access point eMBMS Evolved Multimedia Broadcast [band] API Application programming interface Multicast Services IT Information technology ASA Authorized Shared Spectrum EMF Electro-magnetic field ITU International Telecommunication ATSC Advanced Television Systems EMS Element management system Union Committee eMTC Enhanced Machine Type IWPC International Wireless Industry AWS Advanced wireless services Communications Consortium BBU Baseband unit E-RAN Enterprise RAN KPI Key performance indicator BIU Base station interface unit ETSI European Telecommunications L2ROU Low 2 [watt] ROU BMS Broadcast/Multicast Service Standards Institute LAA Licensed-assisted access BRS Broadband Radio Service E-UTRAN Evolved UTRAN LAN Large-area network BTS Base transceiver station FCC Federal Communications LBT Listen before talk CA Carrier aggregation Commission LEED Leadership in Energy and Cat 5 Category 5 [cable] FDD Frequency-division duplex Environmental Design Cat 6 Category 6 [cable] GPON Gigabit passive optical networks LIPA Local IP access CDMA Code division multiple access GPRS General packet radio service LoRA Long Range [Wide Area Network] CoMP Coordinated multipoint GSM Global System for Mobile LROU Low [power] ROU CPRI Common public radio interface Communications LTE Long Term Evolution CQI Channel quality indicator HD High definition LTE-A LTE Advanced C-RAN Cloud RAN HetNet Heterogeneous network LTE-M LTE for M2M CSI Channel state information HIPAA Health Insurance Portability and LTE-U LTE Unlicensed CTRL Control Accountability Act LWA LTE Wi-Fi aggregation D2D Device to device HROU High [power] ROU M2M Machine to machine DAS Distributed antenna system ICIC Inter-cell interference coordination MAC Media Access Control [layer] DC Data center ICN Information-Centric Networking MCS Modulation and coding scheme DMS Device management system ID Identifier MDU Multiple dwelling units DMS DAS management system iDAS Indoor DAS

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |120| MEC Multiple-Access [was: Mobile] Edge QoE Quality of experience UDN Ultra-dense network Computing QoS Quality of service UDP User Datagram Protocol MIMO Multiple input, multiple output RAN Radio access network UE User equipment MME Mobility management entity RAT Radio access technology UMTS Universal Mobile mmW Millimeter wave RET Remote electrical tilt Telecommunications System MROU Medium [power] ROU RF Radio frequency UTRAN Universal Terrestrial Radio Access MSC Mobile switching center ROI Return on investment Network mUE Mobile user equipment ROU Remote optical unit ViLTE Video over LTE NBI North Bound Interface RPL Third-party logistics VM Virtual machine NB-IoT Narrowband IoT RRC Radio Resource Control VoLTE Voice over LTE NEMA National Electrical Manufacturers RRH Remote radio heads VoWi-Fi Voice over Wi-Fi Association RSRP Reference signal received power vRAN Virtualized RAN NFV Network Functions Virtualization SDH Synchronous digital hierarchy WCS Wireless Communication Services OBSAI Open Base Station Architecture SDN Software-defined networking WDM Wavelength-division multiplexing Initiative SGSN Serving GPRS support node WLAN Wireless local area network oDAS Outdoor DAS SGW Serving gateway OTDR Optical time domain reflectometer SINR Signal to interference-plus-noise OTN Optical transport network ratio OTT Over the top SIPTO Selected IP Traffic Offload PBX Private branch exchange SLA Service-level agreement PCS Personal communications service SON Self-organizing network PDCP Packet data convergence protocol SONET Synchronous optical networking PDN Plesiochronous digital hierarchy SSID Service Set Identifier PHY Physical [layer] TCO Total cost of ownership PIM Passive intermodulation TCP Transmission Control Protocol PON Passive optical network TDD Time division duplex POP Point of presence UC Unified communications

Anritsu, RF interference hunting techniques: Spot, find, fix, Application Note Kathrein, The role of antenna quality in meeting mobile data demand, 2015. 11410-00643, Rev. D, 2014. Levi’s Stadium, Super Bowl 50 at Levi’s Stadium sets event record by transferring Anritsu, Testing distributed antenna systems: S332E Site Master, MT9083 Access 10 terabytes of data on gameday, 2016. Master, MW82119A/B PIM Master, G0306A/B Connector Inspection Lokhandwala, Hatim, Vanlin Sathya and Bheemarjuna Reddy Tamma, Phantom Microscope, Application Note 11410-00887, Rev. A, 2015. cell realization in LTE and its performance analysis, in Proc. of IEEE ANTS, Anritsu, Understanding IBW solutions: In-building wireless: DAS to small cells, 2014. 11410-00885A, 2015. Macknofsky, Gary, Understanding the basics of CPRI fronthaul technology, EXFO, Capretti, C., F. Gringoli, N. Facchi and P. Patras, LTE/Wi-Fi co-existence under 2015. scrutiny: An empirical study, Proceedings of the 10th ACM International Nakamura, Takehiro, LTE enhancements and future radio access, presented at Workshop on Wireless Network Testbeds, Experimental Evaluation & the 5th Future of Wireless International Conference, 2013. Characterization, 2016. NTT DOCOMO, 5G radio access: Requirements, concept and technologies, 2014. China Mobile Research Institute, White paper of next generation fronthaul Paolini, Monica, Boosting coverage and capacity with active DAS: The transition interface, Version 1.0, 2015. from passive DAS to active DAS, Senza Fili, 2015. CommScope, Wireless in buildings: What building professionals think, 2016. Paolini, Monica, Charting the path to RAN virtualization: C-RAN, fronthaul and DeGrasse, Martha, Outdoor DAS and small cells, RCR Wireless, 2016. HetNets, Senza Fili, 2015. Future Mobile Communication Forum, 5G white paper. Paolini, Monica, LTE unlicensed and Wi-Fi: Moving beyond coexistence, Senza Fili, I, Chih-Lin, Corbett Rowell, Shuangfeng Han, Zhikun Xu, Gang Li, and Zhengang 2015. Pan, Toward green and soft: A 5G perspective, IEEE Communications Paolini, Monica, Real-time backhaul assurance to enhance QoE: The evolution in Magazine, 2014. monitoring LTE networks, Senza Fili, 2016. I, Chih-Lin, Shuangfeng Han, Zhikun Xu, Qi Sun, and Zhengang Pan, 5G: Rethink Paolini, Monica, The economics of small cells and Wi-Fi offload, Senza Fili, 2012. mobile communications for 2020+, in Philosophical Transactions of the Royal Paolini, Monica, The smart RAN: Trends in the optimization of spectrum and Society A, 2016. network resource utilization, Senza Fili, 2015. International Telecommunication Union, IMT vision – Framework and overall Paolini, Monica, Voice comes to the fore, again: VoLTE and Wi-Fi Calling redefine objectives of the future development of IMT for 2020 and beyond, voice, Senza Fili, 2016. Recommendation ITU-R M.2083-0, 2015. Piecyk, Walter, What does Verizon’s Super Bowl investment tell us about small Ishii, Hiroyuki, Yoshihisa Kishiyama and Hideaki Takahashi, A novel architecture cells and DAS? BLIG Research blog post, 2016. for LTE-B: C-plane/U-plane split and phantom cell concept, Globecom Workshops conference paper, 2012. Qualcomm, Cost-effective enterprise small cell deployment with UltraSON, 2016.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |122| Rysavy Research, Mobile broadband transformation: LTE to 5G, 5G Americas, SOLiD, The need for speed for drivers and wireless fans alike. 2016. Sutton, Andy, Small cells and heterogeneous networks, TP Journal, vol. 10, part 2, Samsung, LTE radio access solution: Expanding possibilities of network 2016. experience, 2016. Telefonica, in collaboration with Ericsson, Cloud RAN Architecture for 5G. SOLiD, In-building wireless and public-safety imperative, 2015.

REPORT Massively densified networks © 2016 Senza Fili Consulting • www.senzafiliconsulting.com |123| About RCR Wireless News Since 1982, RCR Wireless News has been providing wireless and mobile industry news, insights, and analysis to industry and enterprise professionals, decision makers, policy makers, analysts and investors. Our mission is to connect, globally and locally, mobile technology professionals and companies online, in person, in print and now on video. Our dedication to editorial excellence coupled with one of the industry’s most comprehensive industry databases and digital networks leads readers and advertisers to consistently choose RCR Wireless News over other industry publications. About Senza Fili Senza Fili provides advisory support on wireless data technologies and services. At Senza Fili we have in-depth expertise in financial modelling, market forecasts and research, white paper preparation, business plan support, RFP preparation and management, due diligence, and training. Our client base is international and spans the entire value chain: clients include wireline, fixed wireless and mobile operators, enterprises and other vertical players, vendors, system integrators, investors, regulators, and industry associations. We provide a bridge between technologies and services, helping our clients assess established and emerging technologies, leverage these technologies to support new or existing services, and build solid, profitable business models. Independent advice, a strong quantitative orientation, and an international perspective are the hallmarks of our work. For additional information, visit www.senzafiliconsulting.com or contact us at [email protected] or +1 425 657 4991. About the author Monica Paolini, PhD, is the founder and president of Senza Fili. She is an expert in wireless technologies and has helped clients worldwide to understand new technologies and customer requirements, create and assess financial TCO and ROI models, evaluate business plan opportunities, market their services and products, and estimate the market size and revenue opportunity of new and established wireless technologies. She frequently gives presentations at conferences, and writes reports, blog entries and articles on wireless technologies and services, covering end-to-end mobile networks, the operator, enterprise and IoT markets. She has a PhD in cognitive science from the University of California, San Diego (US), an MBA from the University of Oxford (UK), and a BA/MA in philosophy from the University of Bologna (Italy). You can reach her at [email protected].

© 2016 Senza Fili Consulting, LLC. All rights reserved. The views and statements expressed in this document are those of Senza Fili Consulting LLC, and they should not be inferred to reflect the position of the report sponsors, or other parties participating in the interviews. No selection of this material can be copied, photocopied, duplicated in any form or by any means, or redistributed without express written permission from Senza Fili Consulting. While the report is based upon information that we consider accurate and reliable, Senza Fili Consulting makes no warranty, express or implied, as to the accuracy of the information in this document. Senza Fili Consulting assumes no liability for any damage or loss arising from reliance on this information. Names of companies and products here mentioned may be the trademarks of their respective owners. Cover-page and last-page graphics from brat82/Adobe.