A Survey on VANET Unicast Routing Protocols

www.ijemr.net ISSN (ONLINE): 2250-0758, ISSN (PRINT): 2394-6962

Volume-6, Issue-1, January-February-2016 International Journal of Engineering and Management Research Page Number: 89-94

Mohammad Arif1, Dr. Shish Ahmad2 1,2Integral University, Lucknow, Uttar Pradesh, INDIA

ABSTRACT challenges and perspectives of routing protocols for VANETs MANET is a naive approach for the intelligent are finally discussed. transport system. The routing protocols in VANETs are important and necessary issue. The difference between Keywords--- MANET, VANET, Routing, Protocol VANET and MANET is the mobility pattern and rapidly changing topology. MANETs routing protocols are not effectively applied into VANETs. In this paper, we have made a survey about routing techniques in VANET. We have I. INTRODUCTION introduced unicast protocol for VANET. Carry-and-forward is the new and key concern for designing routing protocols in The growth of the enlarged number of vehicles VANETs. We have also discussed the multi-hop forwarding are equipped with wireless transceivers to communicate and carry-and-forward techniques which reduces the delay. with other vehicles and form a of wireless networks which We have also elaborated the temporary network is known as vehicular ad hoc networks or VANETs [1]. To fragmentation problem and the broadcast storm problem. improve the safety of drivers and provide the comfortable Rapid change in topology causes temporary network driving environment, messages for different purposes need fragmentation problem which influences the performance of to be sent to vehicles through the inter-vehicle data transmissions. The broadcast storm problem affects the rate of message delivery in VANETs. The key challenge is to communications. Unicast routing is a fundamental solve these problems to provide routing protocols with the operation for vehicle to construct a source-to-destination min communication delay, the min communication overhead, routing in a VANET as shown in Fig. 1 and the min time complexity. We have discussed the

Multicast is defined by delivering multicast protocol is utilized for a source vehicle sends broadcast packets from a single source vehicle to all multicast message to all other vehicles in the network as shown in members by multi-hop communication. Geocast routing is Fig. 1 (c). Many results [2-4] on MANETs have been to deliver a geocast packet to a specific geographic region. proposed for unicast, multicast and geocast, and broadcast Vehicles located in this specific geographic region should protocols. However, VANETs are basically different to receive and forward the geocast packet; otherwise, the MANETs, such as the special mobility prototype. This key packet is dropped as shown in Fig. 1 (b). Broadcast differentiation causes the existing routing protocol on

MANETs cannot be directly applied to VANETs. In this transmission delay time is the major concern and the investigation, the recent new results for VANET routing shortest routing path is usually adopted. However, the method are first surveyed. shortest routing path may be not the quickest path with the minimum delay time in VANETs. The shortest routing II. UNICAST ROUTING PROTOCOL path may be found in a low density area, packets can not transmit by the multi-hop forwarding since that there is no This section introduces the unicast routing neighboring vehicle can forward packets. Packets should protocols in VANETs. The main goal of unicast routing in be delivered by carry-and forward scheme. The delay time VANETs is to transmit data from a single source to a is greatly growing if the multi-hop forwarding can not be single destination via wireless multi-hop transmission or utilized. Efforts will be made as finding a routing path with carry-and-forward techniques. In the wireless multi-hop multi-hop forwarding. The min-delay routing protocols [5- transmission technique, or called as multi-hop forwarding, 11] are reviewed as follows. the intermediate vehicles in a routing path should relay data as soon as possible from source to destination. In the 2.1.1 Greedy perimeter coordinator routing protocol carry-and-forward technique, source vehicle carries data as Lochert et al. [5] proposed GPCR (greedy long as possible to reduce the number of data packets. The perimeter coordinator routing) which is a position-based delivery delay-time cost by carry-and-forward technique is routing for urban environment. GPCR protocol is very well normally longer than wireless multi-hop transmission suited for highly dynamic environments such as inter- technique. Two categories of routing protocol designing vehicle communication on the highway or city. GPCR are classified, min-delay routing protocol and traverses the junctions by a restricted greedy forwarding delaybounded routing protocol. Min-delay routing protocol procedure, and adjusts the routing path by the repair aims to minimize the delivery delay-time from source to strategy which is based on the topology of streets and destination. Delay-bounded routing protocol attempts to junctions. Fig. 3 (a) shows that vehicle Vu tries to send maintain a low level of channel utilization within the packets to vehicle VD. Vehicle 1a is selected as the next constrained delivery delay-time. This section describes hop of Vu if greedy forwarding scheme is used. After existing unicast routing protocols in VANETs as follows. vehicle 1a received the packets, vehicle 1a detects 2.1 Min-Delay Routing Protocol destination VD is not located at north. Vehicle 1a then The goal of min-delay routing protocols is to transmit moves packets backward vehicle 2a, then the packet is data packets to destination as soon as possible. The forwarded to VD.

with pre-loaded digital maps, which provide street-level 2.1.2 VADD: vehicle-assisted data delivery routing map and traffic statistics such as traffic density and vehicle protocol speed on roads at different times of the day. Data delivery routing protocol is developed by 2.1.3 Connectivity-aware routing protocol Zhao et al. [6], called as VADD. VADD protocol adopted To overcome the limitation of the static the idea of carry-and-forward for data delivery from a destination, Naumov et al. [3, 7] proposed Connectivity- moving vehicle to a static destination. The most important Aware Routing (CAR) protocol. CAR protocol establishes issue is to select a forwarding path with the smallest packet a routing path from source to destination by setting the delivery delay. To keep the low data transmission delay, anchor points at intermediate junctions. CAR protocol VADD protocol transmits packets through wireless sends the searching packets to find the destination. Each channels as much as possible, and if the packet has to be forwarding vehicle records its ID, hop count, and average carried through roads, the road with higher speed is chosen number of neighbors in searching packets. Once the firstly. VADD protocol assumes that vehicles are equipped searching packets reach the destination, the destination 90 Copyright © 2016. Vandana Publications. All Rights Reserved. www.ijemr.net ISSN (ONLINE): 2250-0758, ISSN (PRINT): 2394-6962 chooses a routing path with the minimum delivery delay toward the destination through the set of anchor points. time and replies it to the source. While destination sends Fig. 4 (a) gives that vehicle VS tries to send data to vehicle the reply packet to the source, the junctions passed through VD, the anchor points are set at I1,1, I2,1, I2,2, I3,2, and by the reply packet are set as the anchor point. After the I3,4. Data is forwarded according to order in the list of path set up, data packets are forwarded in a greedy method anchor points.

2.1.4 DIR: diagonal-intersection-based routing protocol of anchors than CAR protocol [7]. DIR protocol can To improve the CAR protocol, Chen et al. [8] automatically adjust routing path for keeping the lower developed a diagonal-intersection based routing (DIR) packet delay, compared to CAR protocol [7]. protocol. The key difference of CAR and DIR protocols is 2.1.5 ROMSGP routing protocol that DIR protocol [8] constructs a series of diagonal To improve the routing reliability, Taleb et al. [9] intersections between the source and destination vehicles. proposed ROMSGP (Receive on Most Stable Group-Path) The DIR protocol is a geographic routing protocol. Based routing protocol in a city environment. Taleb et al. indicate on the geographic routing protocol, source vehicle that an unstable routing usually occurred due to the loss of geographically forwards the data packet toward the first connectivity if one vehicle moves out of the transmission diagonal intersection, the second diagonal intersection, and range of a neighboring vehicle. In ROMSGP protocol, all so on, until the last diagonal intersection, and finally vehicles are split into four groups based on the velocity geographically reaches to the destination vehicle. For vector. A routing is said as a stable routing if the two given a pair of neighboring diagonal intersections, two or vehicles are categorized in the same group; otherwise, the more disjoint sub-paths exist between them. The novel routing is an unstable routing. A vehicle belongs to a group property of DIR protocol is the auto-adjustability, while if the velocity vector has the maximum projection vector the auto-adjustability is achieved that one sub-path with with this group. Fig. 5 illustrates the ROMSGP routing low data packet delay, between two neighboring diagonal protocol. Two routing paths are established, {VAVB, intersections, is dynamically selected to forward data VBVD} and {VAVC, VCVD}. If VA, VB, VC, and VD packets. To reduce the data packet delay, the route is belong to the same group, the two routing paths are both automatically re-routed by the selected sub-path with stable. Packet is delivered via {VAVB, VBVD} or {VAVC, lowest delay. Fig. 4 (b) shows that DIR protocol constructs VCVD}. If VB turns into another road, the projection a series of diagonal intersections between vehicles VS and vector is changed. VB belongs to the other group. Then the VD. Observe that, DIR protocol may set the fewer number routing path {VAVC, VCVD} is the only choice.

2.1.6 Reliable routing for roadside to vehicle In contrast with routing results developed in the communications highway or the city environments, it is very interest that

Wan et al. [11] specially proposed a reliable routing links. The lifetime of a route is the minimum link lifetime protocol in the rural environment. Wan et al. [11] in a route. Long lifetime of a route improves the routing proposed two reliable routing strategies for roadside to reliability if considered the lifetimebounded shortest path. vehicle (R2V) communication. The challenge of R2V In addition to the lifetime of a routing path, the length- communication in the rural environment is the terrain bounded maximum lifetime path is considered. To factor. For instance, a vehicle moving along the rural construct a length-bounded maximum lifetime path, highway occasionally loses the line of sight (LOS) to the reducing hops can improve the delivery delay-time. A neighbor vehicle or to access points (APs) due to the routing path with fewer hops means the links are obstacle-property caused by the curve roadway and established in the long distance. Establishing a routing path mountains. In addition, almost no fixed communication with longer lifetime implies that the length of this routing infrastructure is available. Multi-hop inter-vehicle path is long. Fig. 6 (a) illustrates the example of lifetime- communication connecting to AP is the main solution of bounded shortest path. The dotted line is current routing the R2V communication. The link lifetime is very path and the link lifetime is going to end, where the important issue for designing the reliable routing. The link minimum link lifetime is 9. The solid line is the candidate lifetime is predicted by two conditions. Once the path. The link lifetime of solid line is greater than the communication is established, the link lifetime halts if (1) threshold (= 16). The routing path changes to solid line by LOS between a pair of vehicles is lost, or (2) one vehicle AP assignment. Fig. 6 (b) illustrates the example of length- moves out of the communication range of the neighboring bounded maximum lifetime path. The dotted line is the vehicle. A link established in a shorter distance usually has routing path with minimum hops to AP (hops = 4). The longer link lifetime. The link lifetime is used to predict the solid line is the selected path (hops = 5). lifetime of a route. A route is constructed by a series of

2.1.7 GVGrid: a QoS routing protocol find vehicle VD and vehicle VD replies RREP message to To improve delivery delay-time and routing VS. Fig. 7 (b) demonstrates that not only the grid sequence reliability, Sun et al. proposed GVGrid protocol [10] but also the information of the next vehicle are recorded in which is a QoS routing protocol for VANETs. GVGrid the routing table. constructs a routing path from source to destination by 2.2 Delay-Bounded Routing Protocol grid-based approach, which divides the map into several Skordylis et al. [12] proposed a delay-bounded grids. The RREQ and RREP packets are delivered through routing protocol in VANETs, which provides a routing different grid to find a routing path through minimum scheme that satisfy user-defined delay requirements while number of grid. A grid is chosen based on the direction at the same time maintaining a low level of channel and the distance between vehicle and intersection and is utilization. The delay-bounded routing protocol [12] selected as next grid if the direction of grid is the same as focuses on the development of carry-and-forward schemes current grid or the grid is closed to the intersection. Then that attempts to deliver data from vehicles to static the intermediate grids between source and destination are infrastructure access point in an urban environment. Two recorded in the routing table. An appropriate vehicle which routing algorithms, D-Greedy (Delay-bounded Greedy has the fewest number of disconnections in each grid is Forwarding) and D-MinCost (Delay-bounded Min-Cost chosen to forward packets to next grid. A formula of Forwarding), evaluate traffic information and the bounded evaluating the expected number of disconnections is delay-time to carefully opt between the Data Muling and derived in [10]. The routing table records in terms of the Multihop Forwarding strategies to minimize source vehicle, destined grid, an appropriate vehicle as communication overhead while satisfying with the delay next hop with minimum the expected number of constraints imposed by the application. D-Greedy disconnections, a vehicle as previous hop, and the grid algorithm adopts only local traffic information to make sequence. Once the routing path is broken, GVGrid just routing decisions. D-Greedy algorithm chooses the shortest finds another vehicle in the grid instead of the previous path to destined AP form the map information, and then vehicle. The routing path dose not require finding again. allocates the constrained delay-time to each street within Fig. 7 (a) shows that vehicle VS floods RREQ message to the shortest path according to the length of streets. If

92 Copyright © 2016. Vandana Publications. All Rights Reserved. www.ijemr.net ISSN (ONLINE): 2250-0758, ISSN (PRINT): 2394-6962 packets can be delivered under the constrained delay-time represents the number of message transmissions in a street. in a street, Data Muling strategy is utilized. Packets are The delay denotes the time required to forward a message carried by a vehicle and forwarded at the vehicle’s speed in a street. To achieve the minimum cost within the to destined AP. Otherwise, Multihop Forwarding strategy constrained delay, DSA (Delay Scaling Algorithm) [13] is is applied if packets cannot be delivered within the applied to select the best routing path with minimum constrained delay-time. Packets are delivered by multi-hop channel utilization under the constrained delay-time. Fig. 8 forwarding. D-MinCost algorithm considers the global shows that Data Muling strategy is applied if the packet traffic information in a city to achieve the minimum can be delivered form VA to AP within the constrained channel utilization within the constrained delay-time. delay-time. Otherwise, the packet is delivered by Multihop According to the global traffic information, the cost and Forwarding strategy. delay of each street can be pre-computed. The cost

2.3 Challenges and Future Perspectives defined delay. Path maintenance shows that a protocol In this section, we have reviewed existing unicast maintains the routing path in passive or active fashion. The routing protocols. Table 1 gives a detailed comparison of passive path maintenance is performed only when a these protocols. Prior forwarding method describes the routing path is expired. The active path maintenance is first routing decision of a protocol when there are packets performed if a routing path is inefficient. To actively to be forwarded. Observe that, delaybounded routing maintain a routing path, the realistic traffic flow protocol is different from other protocols, which carry- information is usually necessary to update the routing path. and-forward is the first considered routing decision to Most of protocols were developed in urban areas under the reserve the wireless media resource. Destination location assumption of high network density. Therefore, some method shows how a protocol discovers the routing path future perspectives should include the following: and destination, which are categorized into two types, 1. A possible future work is how to design min-delay specialized method and integrated method. Specialized unicast routing approaching under low network density. method only indicates the destination location and the The impact of intense density variability should be routing path is discovered while the packet is forwarding. incorporated into the protocol design. The integrated method integrates the path discovery 2. A major challenge in protocol design in VANETs is process into destination finding process. The routing path how to improve reliability of min-delay unicast routing and destination location are simultaneously discovered. protocols to simultaneously reduce delivery delay time and Generally, the integrated method has less routing setup the number of packet retransmissions. time as well as increment of implementation complexity. 3. Driver behavior should be considered for designing of Forwarding strategy is usually in greedy or optimum delay-bounded unicast routing protocols since carry-and- fashion, which expresses what information is considered forward method is the mainly approach to deliver packets. when a protocol establishes the routing path. Greedy 4. To design a routing protocol in a city, the interference forwarding only considers the local information to make by tall buildings along roads should be considered. A the routing decision. Optimum forwarding considers the robustness routing protocol against interference is entire information in a network to choose the best routing appropriate developed in a city environment. path. Normally, optimum forwarding has better 5. Scalability is also an important factor of routing performance; however, lots overheads are required. protocol designing. The VANETs could be a large and Recovery strategy describes route recovery strategy if the metropolitan-scale networks. Protocols should consider routing path is failed. All existing protocols adopt carry- that many unicast routing requests are operating and-forward method except delay-bounded routing simultaneously. The conflict of routing requests between protocol. Delay-bounded routing protocol adopts multi- vehicles should take into consideration, especially in the hop forwarding to reduce the packet delivery time if the intersection. expected packet delivery time cannot satisfy the user-

Table 1: Comparison of unicast routing protocol Protocol GPCR VADD CAR DIR ROMSGP Reliable GVGRID Delay Routing Bounded Prior Wireless Wireless Wireless Wireless Wireless Wireless Wireless Carry Forwarding multi hop multi hop multi hop multi hop multi hop multi hop multi hop forward Method forwarding forwarding forwarding forwarding forwarding forwarding forwarding Destination specialized specialized specialized specialized specialized specialized specialized specialized Location method Forwarding Greedy Optimum Greedy Greedy Greedy Optimum Greedy Greedy Strategy forwarding forwarding forwarding forwarding forwarding forwarding forwarding forwarding Recovery Carry and Carry and Recompute Recompute Carry and Carry and Carry and Multi hop strategy forward forward anchors anchors forward forward forward forwarding Path Passive Active Active Active Passive Active Passive passive maintenance Digital map No Yes Yes Yes No No Yes No needed Realistic No Yes Yes Yes Yes Yes Yes No traffic flow Scenario urban urban urban urban urban rural urban urban

III. CONCLUSION [6] J. Zhao and G. Cao, “VADD: vehicle-assisted data delivery in vehicular ad hoc networks,” IEEE Computer In this paper, we investigated several unicast Communications, 2006, pp. 1-12. routing protocols suitable for VANET environment. The [7] V. Naumov and T. Gross, “Connectivity-aware routing advantages and disadvantages of the surveyed protocols (CAR) in vehicular ad hoc networks,” in Proceedings of are described and the same have been analyzed in terms of IEEE International Conference on Computer packet delivery ratio, end to end delay and latency. Communications, 2007, pp. 1919-1927. [8] Y. S. Chen, Y. W. Lin, and C. Y. Pan, “A diagonal- REFERENCES intersection-based routing protocol for urban vehicular ad hoc networks,” Telecommunication System, Vol. 46, 2010. [1] A. K. Saha and D. B. Johnson, “Modeling mobility for [9] T. Taleb, E. Sakhaee, A. Jamalipour, K. Hashimoto, N. vehicular ad hoc networks,” in Proceedings of ACM Kato, and Y. Nemoto, “A stable routing protocol to International Workshop on Vehicular Ad Hoc Networks, support ITS services in VANET networks,” IEEE 2004, pp. 91-92. Transactions on Vehicular Technology, Vol. 56, 2007, pp. [2] H. Safa, H. Artail, and R. Shibli, “An interoperability 3337-33347. model for supporting reliability and power-efficient [10] W. Sun, H. Yamaguchi, K. Yukimasa, and S. routing in MANETs,” International Journal of Ad Hoc and Kusumoto, “GVGrid: A QoS routing protocol for Ubiquitous Computing, Vol. 4, 2009, pp. 71-83. vehicular ad hoc networks,” in Proceedings of IEEE [3] T. Sawamura, K. Tanaka, M. Atajanov, N. Matsumoto, International Workshop on Quality of Service, 2006, pp. and N. Yoshida, “Adaptive router promotion and group 130-139. forming in ad-hoc networks,” International Journal of Ad [11] S. Wan, J. Tang, and R. S.Wolff, “Reliable routing for Hoc and Ubiquitous Computing, Vol. 3, 2008, pp. 217- roadside to vehicle communications in rural areas,” in 223. Proceedings of IEEE International Conference on [4] R. Manoharan and S. L. P. P. Thambidurai, “Energy Communications, 2008, pp. 3017-3021. efficient robust on-demand multicast routing protocol for [12] A. Skordylis and N. Trigoni, “Delay-bounded routing MANETs,” International Journal of Ad Hoc and in vehicular ad-hoc networks,” ACM International Ubiquitous Computing, Vol. 3, 2008, pp. 90-98. Symposium on Mobile Ad hoc Networking and Computing, [5] C. Lochert, M. Mauve, H. Fera, and H. Hartenstein, 2008, pp. 3017-3021. “Geographic routing in city scenarios,” ACMSIGMOBILE [13] A. Goel, K. G. Ramakrishnan, D. Kataria, and D. Mobile Computing[ and Communications, Vol. 9, 2005, Logothetis, “Efficient computation of delay-sensitive pp. 69-72. routes from one source to all destinations,” in Proceedings of IEEE Conference on Computer Communications, 2001, pp. 854-858.