working papers

availability of improved satellite clocks. astronomical observations. These time tions. In the near future, we can expect Let us now look in more detail at what scales and timing systems result from that discussions among GNSS service Envisioning a Future future clocks might look like. the cooperation of about 60 timing providers and metrologists will begin Importance for TOA-Based Systems. laboratories around the world, which on this subject, and numerous propos- GPS, GLONASS, Galileo, and other continuously contribute to the realiza- als concerning applications and design GNSS System of Systems GNSSes are designed to operate on the tion of UTC. concepts will arise as a result. basis of the principle, “one-way time of UTC rarely differs from the inter- The present achievement of 10- Günter W. Hein, Jose Angel Avila Rodriguez, Stefan Wallner, arrival (TOA) ranging.” Each satellite national average by more than 10 nano- nanosecond accuracy is already an Thomas Pany, Bernd Eissfeller, and Philipp Hartl Part 2 emits its ranging signals together with a seconds. NIST is one of four laboratories astonishing result of many decades of navigation message that tells the user’s worldwide operating the highest prima- research and developments (see Figure receiver from which satellite, We can expect that over the next 1). However, even that achieve- from which orbital position, and ment still leaves much room for at what time it was broadcast. 20 years or so the advances in timing further scientific progress. We By comparing the time of a accuracy characteristic for the past can expect that over the next 20 signal’s arrival with the time of — namely, to improve the stability by a years or so the advances in tim- its transmission, a pseudorange factor 10 per decade — will continue. ing accuracy characteristic for can be calculated. This one-way the past — namely, to improve TOA principle allows an unlimited ry frequency standards to determine the the stability by a factor 10 per decade number of users to use a GNSS. frequency of UTC. — will continue. However, this method assumes that Before we discuss this point further, all GPS satellite clocks involved in a posi- Improved Clocks — So What? it seems appropriate to judge the present tion solution are fully synchronized with What kind of progress and alternative performance from a different perspec- each other and with the International architecture could one invent, if GNSS tive: that 10-nanosecond time resolution GPS time (or some equivalent reference). receivers’ oscillators had a frequency/ corresponds to three meters considering Moreover, it requires extremely precise phase stability orders of magnitudes the propagation of light. At present, the information about the satellite’s position better than the present ones? Moreover, GPS clock stability in space is still of this in a well-defined reference frame. what would be the practical effect if same order of magnitude. Meanwhile, The current standard for all GNSS GNSS satellites could transmit signals the absolute values of the GPS satellites’ For reasons of political sovereignty, technological competition, policy differences, operational control, and satellites is to have three to four clocks for which the uncertainties in GNSS real-time orbital position coordinates perhaps just plain old national prestige, the planet Earth may have four complete global navigation satellite on board each spacecraft — one of them time were not counted in nanoseconds are on the order of one meter. systems within five or six years. Let’s assume that happens. Are users and manufacturers destined to work being the master clock and the others, (with 5 ns corresponding to 1.5 meters These statistics make it quite obvious through a labyrinth of competing technical specifications and management regimes in order to take advantage redundant units. Currently those clocks in ranging error), but rather in femtosec- that many interesting effects of nature of the rich GNSS signal resource coming into existence? Or can we shape a better world of GNSS interoperability are of rubidium and/or caesium types. onds (fs): one billionth of one millionth are still hidden inside the GNSS error and cooperation? Galileo will eventually also use of a second (with 5 fs corresponding to budget, and metrology should aim to a hydrogen-maser clock, if tests on 1.5 millimeter ranging error)? detect and exploit such surprising phe- the second Galileo In-Orbit Valida- It would certainly open a lot of new nomena. Scientists dealing with solid GNSS System of Systems? We is considering adopting it as well. They Innovations in Clocks tion Experiment (GIOVE-B) satellite research topics for scientific applica- Earth research and with fundamental began our discussion in the all predominately use middle Earth If there is an area where revolutionary employing this technology turn out to previous column (January/ orbiting (MEO) satellites with constel- developments could occur in the next be successful. Hydrogen masers have A February 2007 issue) with an lations (current or planned) of between few years, then it is in the field of clocks. superior short-term stability compared exploration of the current status, plans, 24 and 30 space vehicles each. The atomic clocks placed on board the to other frequency standards, but lower similarities, and differences among GPS As we concluded in the last column, satellites are probably the most crucial long-term accuracy due to changes in (United States), GLONASS (Russia), the basis for a GNSS system of systems single element to achieve a high-perfor- the properties of the clocks’ microwave Galileo (), and Com- seems strong, building on infrastructure, mance GNSS. cavity over time. pass (People’s Republic of China). operations, and policies that are already Because the development cycle in Atomic clocks are also installed at Consider the similarities: All four in place or under development. Now we clock technology is about 7 years, these ground stations of those GNSS systems. are global systems accessible by users will turn to some concepts further out- newest generation technologies might be GPS clocks in space and on the ground worldwide. They have one or more radio side the realm of the expected — some available on orbit in a timeframe of 20 are in some way synchronized to the frequency bands in common where they speculative possibilities based on inno- years. This would considerably alleviate International Atomic Time (TAI), which broadcast open signals free of charge. vative ideas, but which nevertheless are the challenge of generating predicted is the world’s continuous and stable time They have comparable atomic time and within the realm of the possible. satellite clock corrections. Furthermore, scale. Alternatively (or additionally), the geodetic coordinate frames. And nowhere is there more room one might speculate that, since orbit pre- clocks are also tied to the civil Coordi- Three of them have a common sig- for innovation than in the domain diction already works quite well today, nated Universal Time (UTC). Derived FIGURE 1 Accuracy of terrestrial caesium atomic clocks at the National Institute of Standards and nal technology — code division multiple that underlies GNSS technology itself: differential correction service for GNSS from TAI, but synchronized with the Technology. The NIST-7 is an optically pumped, thermal atomic-beam, microwave caesium spec- trometer and the NIST-F1 is a caesium fountain atomic clock access (CDMA) — and the other system time. systems could become obsolete with the passing of day and night on the basis of

64 InsideGNSS march/april 2007 www.insidegnss.com www.insidegnss.com march/april 2007 InsideGNSS 65 working papers

The two very different types of instru- Atomic Clocks: ments will be synchronized by means Not Just for Satellites of a specifically developed “Frequency Improved timing in GNSS receivers is Comparison and Distribution Package” technically already possible today. How- (FCDP), and the results shall be trans- ever, the development of good, stable mitted to ground by two datalinks, a and at the same time extremely cheap laser- and a Ku-band link. reference oscillators has not kept up This experiment will distribute a with progress achieved in other aspects stable and accurate time base for space- of GNSS receiver chip design. Conse- to-ground and ground-to-ground time quently, all types of GNSS receivers still and frequency comparison. Institutes use crystal oscillators for their timing worldwide will participate in comparing reference. their ground-based atomic clocks with Consider, for example, one very the ACES clock signal. Analysis of the interesting development: in 2004 the scientific data will seek to shed light on U.S. National Institute of Standards fundamental relativistic issues, as well as and Technology (NIST) demonstrated FIGURE 2 Color scheme of femtosecond laser frequency comb-synthesizer (Courtesy of Prof. T. W. for time and frequency metrology, geod- a chip-scale atomic clock (CSAC) about Hänsch. For more details refer to the Nobel Lecture in the Additional Resources section) esy, gravimetry, precise orbit determina- the size of “a grain of a rice” and stable tion, and Earth monitoring by means of enough at 10-9. The CSAC’s volume is relativistic issues, for example, offer ground-based clocks will become avail- the very long baseline interferometry less than 10 cubic millimeters (1.5mm highly interesting possibilities for fur- able for space applications a few years (VLBI). × 4mm) with power requirements for FIGURE 3 Atomic Clock Technology Progression of CSACs. Figure courtesy of DARPA by Clark Nguyen ther accuracy improvements. later with approximately the same per- The ACES project may also provide battery driven applications of 75 mil- (see Additional Resources citation for more details) Certainly, users dealing with prac- formance qualities. insight into the question of whether, in liwatts. tical applications will follow. We can In fact, scientists emphasize that the far future, the world’s timekeepers A complete atomic clock based on anticipate this likelihood from the expe- atomic clock in space will actually would be better off using a space-based the CSAC would be about one cubic world that contribute to the determina- CSAC development succeeds, the target riences of the past: In the 1980s, when eliminate some of the perturbations that time and frequency reference system as centimeter in size according to NIST. tion of the UTC. timing accuracies of 10-11 with stability intercontinental time synchronization generate problems for fountain clocks on the primary standard or retaining the (For details, see .) hydrogen maser and four caesium-beam in small devices of only 1 cubic centime- second and better timing resolution was the Earth’s magnetic field, and other In this aspect, the GNSS com- This development alone could stim- clocks. It is as accurate that it would nei- ter with power consumptions of only 30 requested from scientific community, phenomena provide advantages for the munity should also investigate which ulate new inventions and alternative ther gain nor lose one single second in milliwatts. almost no “practical user” was interested performance improvements. other phenomena might play a role in designs for GNSS receivers. Such prod- more than 60 million years! Having such accurate clocks the in such an improvement. Nowadays the For example, on Earth, gravity causes combination with time and distance. ucts might not compete with the vari- size of a wristwatch would open a new nanosecond technique is standard in an atomic clock’s caesium atoms to fall For example, space–based VLBI could ous existing and successfully used GPS Chip-Scale world of amazing applications. The clear metrology! away from the detector almost imme- prove to be extremely important for receivers of the civil mass market, but Atomic Clocks motivation for CSACs is to enable ultra- As one practical advantage of an diately. This makes it more difficult to the definition of coordinate reference they could instead lead to new develop- As shown by Bill Klepczynski during miniaturized and ultra–low power time improved time and frequency refer- determine the center of the atom’s oscil- frames, for UTC issues, and for astro- ments in science and technology as well the 46th CGSIC meeting on Septem- and frequency references and broaden ence, system operators might be able to lations. In microgravity caesium atoms physics in general. as applications in professional, commer- ber 26, 2006, atomic clock technology the associated applications as much as dispense with satellite clock corrections move 5 to 10 times slower. Femtosecond Lasers. Optical frequen- cial, and military realms. is progressing very quickly. As we see possible. entirely. This could also reduce the need PARCS and ACES. NASA’s Primary cy standards have been based on laser- Today GPS is already often combined in Figure 3, the state of practice at the At the moment, five groups are of corrections in the orbit prediction to Atomic Reference Clock in Space cooled atoms and ions for a long time with other sensors such as inertial mea- moment stems from the NIST-F1 pro- working on this promising technology very few cases, which would simplify the (PARCS ) program intends to demon- (see Figure 2). These systems promised surement units or dead reckoning devic- totype. With accuracies on the order divided in two main programms. In the whole issue of GNSS control. strate such influences with its experi- superior stability over existing micro- es. Because such integrated solutions are of 3.8×10-15 and stability values rang- Defense Advanced Research Projects mental package on board of the External wave standards. However, dividing often the only possibility in poor GNSS ing 3.3×10-15/hr, the physical dimen- Agency (DARPA) program we can find Space-Borne Facility of the Japanese Experimental down the very fast optical oscillations to signal environments to allow position- sions and power consumption are not NIST, Symmetricom, Honeywell, and Atomic Clocks Module section of the International a countable frequency has proved to be ing, we expect that this trend will gain so encouraging because a volume of Rockwell Scientific while in the Office The atomic clocks on board GNSS satel- Space Station (ISS). Meanwhile, the ESA extremely complicated until recently. even more in importance. 3.7 cubic meters and power consump- of Naval Research program, Kernco is lites are space-qualified samples of ter- project ACES (Atomic Clock Assembly New frequency dividers based on Development Trends of Atomic Stan- tion of 500 watts are currently needed. the main actor. Figure 4 shows the CSAC restrial units. Specific design features of in Space), planned to start on the ISS in mode-locked femtosecond lasers and dards. Since the U.S. National Bureau Such characteristics are impractical for concept. the spaceborne clocks take into account 2010, will also study the benefits that microstructure optical fiber provide of Standards (the predecessor of NIST) receiver applications. One of the main and most interest- the well-known effects and constraints might result from future space-borne convenient, robust, accurate means of developed the first atomic clock in 1949, Nevertheless, although this is the ing applications for CSACs would be to concerning the environmental condi- clocks. phase-coherently linking optical fre- the frequency stability and the time sta- current state of the art, the atomic attain rapid re-acquisition. Indeed, as we tions, power and weight problems, and Two of the most modern atomic quencies to standards in the microwave bility have been improved every decade clock technology progression is moving know, the choice is between more cor- so forth. However, in principle, one can clocks, the space hydrogen maser (SHM) domain. This breakthrough has opened by one order of magnitude, as shown in quickly toward significant reductions in relators and a good clock and as shown assume that the development results of and the cold caesium-clock shall be test- the door to a new generation of atomic Figure 1. NIST is one laboratory among the required dimensions. CSACs could by Klepczynski, reducing the frequency the frequency standard institutes for ed in the microgravity field of the ISS. clocks based on optical transitions. the 60 timing laboratories around the be a reality in the foreseeable future. If instability from 10-7 to 10-11 would be

66 InsideGNSS march/april 2007 www.insidegnss.com www.insidegnss.com march/april 2007 InsideGNSS 67 working papers

Europe and the United States have against all sources of potential interfer- for tracing global atmospheric humidity some major aspects that could help agreed on broadcasting the Galileo-GPS ence, all systems are expected to finally maps, contributing to improved global improve reception time offsets mutually. adopt CDMA in the future. weather forecasting. in signal-challenged environments. If GLONASS and the upcoming New RNSS bands? The radionaviga- Multi-Mission Concept? Satellite mis- For instance, increasing the transmit- GNSS systems, including augmenta- tion satellite service (RNSS) portion sions are expensive, and consequently ted signal power would definitely help. tions, follow this approach (for example of the RF spectrum is overcrowded. the trend is to put many missions togeth- One could imagine that it might be pos- QZSS plans to do so), we can also con- This is especially true on the L1 band. er. If we take a look at GPS III, we can sible to increase satellite signal power by sider the issue of a common time refer- Nevertheless even those bands that see proposals of equipping the satellites only a few decibels. A further increase ence system to have been resolved. One have not been used yet will certainly with weather sensors, nuclear detectors in received signal power might become just has to keep in mind that the number be shared by many systems in the near (already on board existing generations possible using directional transmitting of broadcast offset values increases as future. Thus, the search of other free of GPS satellites), earth observation, antennas (similar to the GPS spot beam the square of the number of compatible frequency resources is something that and atmospheric sensors. The objective mode planned for military signals) giv- systems operating under such reciprocal will occur with a high probability in the is to extend the payload capability to the ing, perhaps an additional 20 decibels of agreements. next years. maximum to spread costs across mul- received power. If future ground segments had multi- As we know, the Galileo program tiple missions, and the trend will remain Another idea would be to choose a system navigation receivers installed at obtained authorization to use a C-band in the future. frequency that is better suited than an L- monitoring sites and reference stations, assignment for radionavigation. The For some time QZSS was expected to band signal to penetrate buildings. One FIGURE 4 Anatomy of a chip-scale Atomic Clock. Figure courtesy of DARPA by Clark Nguyen (see Ad- they could make a great contribution to technical complexities, however, have offer a communication service together recalls, for example, that Loran-C sig- ditional Resources for more details) the very precise definition and subse- made it impossible for the first genera- with navigation. Although the idea nals are known to provide rather good quent maintaining of a global time and tion of Galileo to take advantage of the seems to have been abandoned, similar reception inside buildings. However, possible using CSACs. The result would Time, Coordinates, coordinate reference frame. C-band allocation. approaches could be revisited again in Loran-C is also known to exhibit differ- be a reduction of the Time To First Fix and Orbits Indeed, phase noise problems, the coming years. ent problems regarding interference, and (TTFF) from 93 seconds to only 5 for a From the users’ perspective, common Satellite Operations & higher free space attenuation (related to Additionally, GNSS could also be a space-borne signal might not be able to similar correlator size of 6000 correla- GNSS time and coordinate reference sys- Opportunities the use of omni directional antennae) used not to only position on the ground penetrate through the ionosphere. tors. This means that the improvement tems are needed to simplify the simul- Satellite operations — signals, frequen- and the strong signal attenuation due but also in space. Why not to use On the other hand, many signal brought by good clocks is equivalent to taneous use of different GNSSes. For cies, payloads, communications — offer to rain knocked down all the proposed GNSS to position geostationary satel- reflections occur within buildings; so, that of increasing the number of cor- political and technical reasons, however, a number of opportunities for coopera- solutions. But in some decades, things lites? Today GEOs are placed into their a wide bandwidth signal might help relators, regardless of the real costs of the various GNSS systems should not be tion among the GNSS system operators could have changed; thus could maybe intended orbital locations with accura- separate direct and multipath signals. both. In fact, new CSACs would have to overly similar or depend on each other. to improve the collective robustness of a the C band be an alternative? What cies of around 1 degree. But if GNSS One could imagine that the satellites be priced very low to compete against Given this premise, the most that system of systems while greatly improv- about adding lower frequencies to con- were used here, too, the tolerance cold be broadcast ultrawideband (UWB) signals correlators. seems achievable is use of a common ing the users’ experience. tribute towards a more effective indoor reduced to lower values of for example for that purpose. This involves frequen- “Nugget” of Synchronization. The standard to define the reference systems, CDMA or FDMA? GPS, Galileo, and use of GNSS? We will return to this 0.2 degrees, thus allowing for a denser cies in the range of 3-10 GHz. Naturally, integration of chip-scale atomic clocks but allowing each GNSS to employ dif- China’s planned Compass system employ subject in Part 3. There, we will depict network of geostationary satellites and a the code pseudorange noise would be into new GPS receivers is by no matter a ferent realizations (frames) of the given code division multiple access (CDMA) an imaginary scenario where C-Band more efficient exploitation of the limited extremely small combined with excel- far-future technology. Last year the U.S. standard (system). This goal has been techniques in their systems’ design. On would be reserved for military/govern- space resources in that specific orbit. lent multipath mitigation capabilities. Navy Space and Naval Warfare Systems achieved between GPS and Galileo. Both the other hand, GLONASS has used and mental applications leaving the L band Apart from these rather hypothetical Center San Diego announced their broadcast satellite positions in a refer- still uses frequency division multiple alone for civil users (or vice versa?). This Dedicated Indoor-GNSS? options, a future GNSS system should “Navigation Nugget” receiver (Inside ence frame based on the International access (FDMA) technology. would have very interesting benefits to The present and planned GNSSes are GNSS, April 2006). Terrestrial Reference System (ITRS). As long as a common technique is both types of user and important con- definitely the primary positioning The Navigation Nugget consists of a Slight coordinate differences at the not used by all GNSS services, we can- sequences. resource when line-of-sight signals from Erratum Inter-satellite Links? The authors would like to mention that in the software-defined GPS receiver and an millimeter level might exist due to the not say that the systems will be fully Inter-satellite the satellites can be received. Over the January/February issue the two following inertial measurement unit (IMU), both use of different realizations of the same interoperable, as users would desire. links could be something quite common last decade product designers and man- erratas were found: coupled to the onboard atomic clock. standard, but those differences are not During the GPS/GLONASS Working in not so many years. The Galileo pro- ufacturers have considerably increased 1. As noted by Grace Pazos made us note, the The design enables highly precise posi- significant for the navigation user. In Group 1 meeting in December 2006, gram has studied the possibility but for GPS receiver sensitivity (by nearly 30 first GPS Block IIR-M (SVN 53/PRN 17) was tioning and navigation also in highly contrast, GLONASS coordinates are both sides emphasized the benefit to financial reasons did not employ them dB). Today satellite navigation even launched on September 26, 2005. December impaired and threatened GPS terrain. based on the PZ-90 reference system, the worldwide GNSS user community a in the first generation. GPS has carried works with attenuated signals, for exam- 16, 2005, was actually the date it was set usable for operational purposes. Coupling of GPS, IMU and atomic making the coordinate conversion more common approach concerning FDMA/ out first tests with inter-satellite links. ple, inside a vehicle’s glove compartment 2. As correctly noted by Christian Tiberius, clocks is going to assist users in position- cumbersome. CDMA would bring. The Russian side GLONASS and GPS will implement or on the upper floors of buildings. Table 3 mentions 64 E for one of the GEO- ing in jammed and shaded environments Similarly, all time systems of GPS, announced that they will come to a them on one of the next generations. Notwithstanding those facts, GNSS longitudes of one of the EGNOS satellites. as well as in the critical area of transition GLONASS, and Galileo are based on decision about possible addition of or Without doubt, for achieving shorter is definitely still not a dedicated or In fact, this is the old ESTB satellite (PRN between indoor and outdoor. A Naviga- UTC, but using individual realizations conversion to CDMA technology by the times-to-alarm for integrity purposes, autonomous means for indoor position- 131) and the current EGNOS PRN 126 is much closer to Europe, at 25 E. tion Nugget of enlarged scale is expected of UTC. “Spending” one satellite’s obser- end of 2007. inter-satellite links are the only way to go. ing. Can this performance be improved Moreover, we would like to thank Christian to be field-tested within months. vations enables a GNSS receiver itself to CDMA technology allows for a great Moreover, one can even envision these and, if yes, how? Tiberius and Grace Pazos for their helpful solve for the time offset between two separation between signals that share the links supporting applications of atmo- Although we have no definitive comments in finding these errors. different satellite systems. Up to now same band. Given its great robustness spheric sounding (GNSS occultation) answer about this, we can point out

68 InsideGNSS march/april 2007 www.insidegnss.com www.insidegnss.com march/april 2007 InsideGNSS 69 working papers

extremely simple to operate. A local ser- age might no longer need a regional aug- out in the past years to assess the effect of cooperation between the United States tem itself, and GPS III will follow this vice provider can place them relatively mentation, and then the natural question increasing the number of satellites of an and the European Union an interference path as well. At least the two will have easily in locations where satellite navi- arises: what is the added value of using existing constellation in which the effect compatibility methodology was devel- it. GLONASS might embrace the idea, gation signals can not be received. Just regional systems when all the planned of doubling the number of satellites was oped in conjunction with the NSCC too, and possibly Compass as well. imagine a pseudolite in the form of a global systems are already delivering the studied in detail. In terms of position- following ITU standards. The bilateral The only problem is that the integrity light bulb operating completely autono- accuracy that we need? ing accuracy, the improvement result- Agreement, however, is only between concepts developed — or under develop- mously. Of course, this area requires a ing from the better geometry is obvi- Americans and Europeans. If new sys- ment — for Galileo and GPS are differ- lot of standardization work in order to Are so many satellites ous. Indeed, the step from GPS alone to tems come into play, standardization ent. As a result we cannot strictly talk control the mutual interference of pseu- really necessary? Galileo + GPS undoubtedly represents a will be needed and perhaps even multi- about global integrity as long as no har- FIGURE 5 Qualitative analysis of the expected marginal gain as a function of the number of dolites and GNSS signals and to allow Having as many satellites as possible is clear gain for the final user. lateral agreements. Otherwise chaos will monization exists among the methods. GNSS satellites GNSS receiver manufacturers to process always good. Nevertheless, every extra Nevertheless once a reasonably reign in the RNSS band and the law of What happens, then, if every system those signals. satellite implies an important extra cost dense constellation of satellites is the strongest will prevail. has different integrity requirements and to society and therefore we might well achieved, would not the user profit more Who should and could be respon- concepts? Should Galileo assure a six- GNSS beyond the earth ask if a point is reached at which no real from an additional increase of power sible for coordinating these actions? It second integrity constraint but the oth- for space exploration? improvement can be observed when the than from an increase in the number seems that the United Nations Office of ers do not, then what use was it for Gali- Why should we limit ourselves to size of the combined GNSS constellation of satellites? Let us think again of our the Earth? We hear everywhere increases. future scenario of 110 satellites. What Satellite navigation is fashionable. Every superpower about U.S. plans to return again Indeed, one might expect that the rel- would be better for the user: one extra wants to have a satellite navigation system of its own to the moon and start from there ative gain brought by 30 satellites when satellite or an increase in the transmit- and preferably a global one. with the human exploration of there are already 30 existing is higher ted power? other planets in the solar system. than that of 30 additional satellites when The problem gains even more multi- Outer Space Affairs could play such a leo to have a more stringent design than Putting antennas point- there are already 60. This is equivalent to dimensionality if we recall the develop- role — extending its efforts in sponsor- the others? Can we talk about integrity ing towards the space is some- saying that the marginal gain diminish- ment of the semiconductor industry in ing formation of the International Com- at all if each GNSS understands it a dif- thing that could become a com- es as the size of the constellation grows. recent years. As mentioned earlier, it is mittee on GNSS (ICG). But transforming ferent way? mon practice within only a few Such a conclusion should not come as a not so unrealistic to think that not too existing bilateral agreements into com- Additionally, the concept of integrity decades. GNSS would then still surprise, because it reflects a well known far in the future pseudolites could be a patible multilateral agreements is not an itself is also pretty open. We have the give us a position and a time, but economic law that applies to most goods cheap product that anyone could place easy task. For better or worse, the more standard integrity defined for the avia- what time? With respect to the and services in the world. in locations with poor GNSS coverage. players are involved, the more difficult it tion community (by the International earth? In a lunar or solar coordi- As reflected in Figure 5, the four These single-chip pseudolites (SCPL) has always been to agree on things. Civil Aviation Organization or ICAO), nate system? global satellite systems now in existence could thus meet users’ positioning GPS and Galileo are compatible and another for maritime, land, etc. The Perhaps we already should or under development will have around requirements in areas where satellites interoperable to a high degree, but are different understanding from different begin thinking about implement- 110 satellites altogether. Do we really signals could not be received, no matter not equal. Although important common communities should flow in a structured ing solar system time such as a need so many? To what region of Figure how dense the network of satellites. actions have occurred in recent months, way so that in the future all the systems NASA speaker announced in the 5’s curve do 110 satellites correspond? there is still a long way to go. And the would be certified for use worldwide in FIGURE 6 Example of GNSS satellite density could be 2006 Munich Satellite Navigation An imaginable possibility to estimate Standardization difficulties are compounded when we whatever mode or operating environ- selected Summit. the saturation level in terms of geom- and Harmonization compare both systems with GLONASS ment, and all the systems should incor- etry would be to calculate the number Satellite navigation is fashionable. Every or the planned Compass. porate certified RAIM. generally provide an optimized signal The GNSS System of satellites that are needed to cover a superpower wants to have a satellite No matter what the system designers For other life-critical services as structure for high sensitivity receivers to of Systems difficult urban canyon environment, navigation system of its own and pref- do, the fact is that the user market will search and rescue common frequencies allow easy acquisition and reliable track- As we have seen from our previous dis- such as shown in Figure 6. Indeed, if erably a global one. But, is there room explode in the next years. GNSS receivers would be desirable as well as compatible ing. For example an intelligent mixture cussion in the January-February 2007 we assume a house of 25 meters height for everyone? We saw in the previous will work better and almost everywhere, modulation and code rates. between short PRN codes for easy acqui- issue of Inside GNSS, four global naviga- with a distance of 10 meters between two discussion that on the other hand the and the fusion with other communica- sition and long PRN codes in case when tion systems might be fully operational buildings, an angle of approximately 12 real need of having multiple systems is tion devices is already on the threshold. Conclusions cross-correlations are important would within two decades providing, an excel- degrees results with respect to the satel- open to question at some point from an International standards and certification GNSS is already a success and, as hap- be desirable. Furthermore, it should pro- lent coverage of most of the locations on lites and the middle of the street. If we economic perspective. are urgently needed and although it is pens with any success, everybody wants vide long-term predictions for satellite the earth. further assume the same semi-major A very different issue, but one of great true that ICAO, ITU, RTCA already pro- to have part of the cake. The GNSS Sys- orbit and clocks. Today, with GPS alone an average of axis as Galileo and five MEO planes, importance, is whether we can have so vide models for certification and regula- tem of Systems of the future will be the Satellite navigation will also do a approximately 10 satellites can be seen all the Earth could be covered for eleva- many systems coexisting together with- tion, the market forces are stronger and result of putting together many differ- good job in helping other indoor navi- at any point of the earth. When GPS, tion angles higher than 78 degrees by 5 out degrading the performance of one demand faster reactions. ent navigation systems from different gation systems. Among the latter, ultra- GLONASS, Galileo, and COMPASS are × 22 (or 110) satellites. This number, of another. The interference caused on one countries. tight GPS/INS integration, wireless net- operational — assuming they would be course, corresponds to about four GNSS system by the rest is technically difficult Global vs. Regional Integrity Nevertheless, the final users are only work positioning, RFID, and terrestrial fully interoperable — four times more constellations, whose orbits are opti- to measure, and especially if each system Everybody talks about integrity — the interested in receiving signals — no mat- UWB, are promising candidates. satellites may be available for navigation, mized with respect to each other. would develop its own concept without ability to warn users about “unhealthy” ter where they come from or to whom The concept of pseudolites also could positioning, and timing. In the framework of the Galileo pro- taking the rest into account. — erroneous — satellite signals. Galileo they belong. Thus a true global system be of interest, if they are cheap and Locations that now have poor cover- gram, various studies have been carried In the 2004 agreement on GNSS will have integrity built into the sys- of systems would be with no doubt his

70 InsideGNSS march/april 2007 www.insidegnss.com www.insidegnss.com march/april 2007 InsideGNSS 71 working papers

GPS Solutions, Volume 8, Number 3, 119-139, the Galileo signal structure, GNSS receiver desire. Today we can only imagine what September 2004, Wiley Periodicals Inc., 2004 design and performance, and Galileo codes. it will look like, but the shape of that [15] Irsigler, M. and G. W. Hein, B. Eissfeller, A. Bernd Eissfeller is a full future system will depend on the deci- Schmitz-Pfeifer, M. Kaiser, A. Hornbostel, and P. professor of navigation sions that we make today. Hartl, ” Aspects of C-Band Satellite Navigation: and vice-director of the The time for cooperation, harmo- Signal Propagation and Satellite Signal Tracking” Institute of Geodesy and nization, and standardization is here. In: EUGIN [Hrsg.]: Proceedings on CD, ENC GNSS Navigation at the Nobody says that it will be an easy task, 2002, May 27–30, 2002, Kopenhagen University FAF Munich. He but wouldn’t it be a great thing to have [16] IS-GPS-200: NAVSTAR GPS Space Segment is responsible for teaching and research in receivers in the near future that can /Navigation User Interfaces [17] Nguyen, C. & Kitching, J. “Towards chip- navigation and signal processing. Till the end of receive dozens of satellites in difficult 1993 he worked in industry as a project manager scenarios where no single system alone scale atomic clocks,” in Digest of Technical Papers, ISSCC 2005, 1, 84-85, Piscataway, NJ: on the development of GPS/INS navigation can offer reliable services today? IEEE, 2005. systems. He received the Habilitation (venia legendi) in Navigation and Physical Geodesy in [18] Nobel lecture: Passion for Precision (T. W. 1996 and from 1994-2000 he was head of the Additional Resources Hänsch). Review of Modern Physics 78 (2006) GNSS Laboratory of the Institute of Geodesy and [1] Avila-Rodriguez, J.A. et al. (2005), “Revised 1297-1309 Navigation. Combined GALILEO/GPS Frequency and Signal [19] Revnivykh, S., “GLONASS Status Update”, Thomas Pany has a Ph.D. Performance Analysis”, Proceedings of ION Proceedings of 46th CGSIC Meeting, 25 in geodesy from the Graz GNSS 2005 – 13-16 September 2005, Long September2006, Fort Worth, Texas, USA Beach, California,USA. University of Technology [2] Avila-Rodriguez, J.A et al. (2006), “How Authors and a M.S. in physics from to optimize GNSS signals and codes for indoor the Karl-Franzens “Working Papers” explore University of Graz. positioning”, Proceedings of ION GNSS 2006, the technical and scien- 26-29 September 2006, Fort Worth, Texas, USA Currently he is working at tific themes that underpin the Institute of Geodesy [3] China Daily, Beijing, 13 November 2006 GNSS programs and appli- and Navigation at the University of Federal [4] Galileo SIS ICD 23/05/2006 cations. This regular col- Armed Forces Munich. His major areas of interest [5] Hein, G.W., Eissfeller, B. (2001), umn is coordinated by include GPS/Galileo software receiver design, “Technology Assessment for Precision Prof. Dr.-Ing. Günter Galileo signal structure and GPS science. Positioning and Navigation Scenarios 2005 and Hein. Stefan Wallner studied at 2010 for Automotive Applications,” prepared Prof. Hein is a member of the European Com- the Technical University for DaimlerChrysler Research and Technology, 9 mission’s Galileo Signal Task Force and organizer of Munich and graduated March 2001 of the annual Munich Satellite Navigation Sum- with a diploma in techno- [6] Hein, G. W. et al. (2006), “MBOC: The New mit. He has been a full professor and director of mathematics. He is now Optimized Spreading Modulation Recommended the Institute of Geodesy and Navigation at the research associate at the for GALILEO L1 OS and GPS L1C”, Proceedings University of the Federal Armed Forces Munich Institute of Geodesy and of ION PLANS 2006, 24-27 April 2006, Loews (University FAF Munich) since 1983. In 2002, Navigation at the Coronado Bay Resort, San Diego, California he received the United States Institute of Navi- University FAF Munich. Wallner’s main topics of [7] http://de.wikipedia.org/wiki/Atomic_Clock_ gation Johannes Kepler Award for sustained and interests are the spreading codes and the signal Assembly_in-Space significant contributions to the development of structure of Galileo and also interference and [8] http://gps.losangeles.af.mil/engineering/ satellite navigation. Hein received his Dipl.-Ing interoperability issues involving GNSS systems. icwg/Docs/EC%20 and%20US%20Joint% and Dr.-Ing. degrees in geodesy from the Univer- Phil Hartl has an M.S. in 20Statement%20on%20GALILEO%20and sity of Darmstadt, Germany. Contact Prof. Hein at electrical engineering and %20GPS%20Signal%20Optimization%20- . a Ph.D. Habil. in space %20%2024%20Mar%2006.pdf José-Ángel Ávila- technology from the [9] http://gps.losangeles.af.mil/engineering/ Rodríguez is a research Technical University of icwg/Docs/WGA%20Signed%20Recommendatio associate at the Institute Munich, a Ph.D. in n%20on%20MBOC%20-%2023%20Mar%2006. of Geodesy and aeronautics and Pdf Navigation at the astronautics from the University of Stuttgart. He has served as the head of the DLR Space Research [10] http://pnt.gov//public/docs/2004-US- University FAF Munich. Centre Oberpfaffenhofen, director of the Institute ECagreement.pdf He is responsible for research activities on of Satellite Technology of the Technical University [11] http://tf.nist.gov/ofm/smallclock/index. GNSS signals, including BOC, BCS, and MBCS of Berlin, and director of the Institute of htm modulations. Ávila-Rodríguez is involved in the Navigation of the University Stuttgart. He was [12] http://www.glonass-ianc.rsa.ru/i/glonass/ Galileo program, in which he supports the principal scientists for many satellite joint_statement_eng.pdf , the European experiments and numerous national and [13] http://www.space.com/spacenews/ Commission, and the Galileo Joint Undertaking, international projects. Since his retirement as archive05/Nigeria_0221.html through the Galileo Signal Task Force. He studied full professor in 1995, Dr. Hartl has been active as [14] Irsigler, M., and G. W. Hein and A. Schmitz- at the Technical Universities of Madrid, Spain, scientific consultant to various research and Peiffer (2004): Use of C-Band frequencies for and Vienna, Austria, and has an M.S. in electrical industrial organizations in various fields of satellite navigation: benefits and drawbacks, engineering. His major areas of interest include aerospace.

72 InsideGNSS march/april 2007 www.insidegnss.com