performance comparison of position-based routing protocols in vehicle ...

Ram Shringar Raw et al. / International Journal of Engineering Science and Technology (IJEST)

PERFORMANCE COMPARISON OF POSITION-BASED ROUTING PROTOCOLS IN VEHICLE-TOVEHICLE (V2V) COMMUNICATION RAM SHRINGAR RAW, SANJOY DAS School of Computer & Systems Sciences Jawaharlal Nehru University New Delhi, India Abstract: Vehicular Ad hoc Networks (VANETs) is the new wireless networking concept of the wireless ad hoc networks in the research community. Vehicle-to-Vehicle (V2V) communication plays a significant role in providing a high level of safety and convenience to drivers and passengers. Routing in VANET is a major challenge and research area. Position based routing protocol has been identified to be suitable for VANETs because of frequently changed network topology and highly dynamic nature of vehicular nodes. Many position based routing protocols have been developed for routing messages in greedy forwarding way in VANETs. However, few of them are efficient when the network is highly dynamic. In this paper, we present an overview and qualitative comparison of existing position based routing protocols that are based on the position prediction of neighboring and destination nodes. We evaluate the performance metrics such as end-to-end delay and packet delivery ratio using NS-2 simulator. Keywords: VANET; V2V; Routing Protocols; Position-Based Routing Protocols; AMAR; GyTAR; EBGR; BMFR. 1. Introduction Vehicular traffic accidents on the road cause losses of thousands of lives, injuries and huge material damages every year. Traffic rules violations are the main reasons of the vehicular traffic accidents. Therefore, having an efficient way to detect violations will yield reductions of traffic accidents and enable efficient traffic management system. Recent advances in telecommunications, computing and sensor technology emerged the vehicular environment as one of the hottest research areas for the communications industry. To reduce large number of vehicular traffic accidents, improve safety, and manage traffic control system with high and reliable efficiency, computer networking researchers have proposed a new wireless networking concept called Vehicular Ad hoc Network (VANET) which can increase passenger safety and provide “efficient” road and policies monitoring. In the future, VANET will provide safer and well-organized road and a large number of vehicular applications ranging from transport automation systems to entertainment and comfort based applications [Papadimitratos (2008)]. VANET [Olariu (2009)] is one kind of vehicular communication based on wireless network technology to establish the wireless ad hoc network between vehicles (see fig. 1). In 1999, the Federal Communication Commission (FCC) [Abdalla (2008)] allocated a frequency spectrum for vehicle to vehicle (V2V) and vehicle to roadside (V2R) wireless communication. The Commission then established the Dedicated Short Range Communications (DSRC) service in 2003. DSRC is a communication service that uses the 5.850-5.925GHz frequency band (5.9 GHz band) for the use of public safety and private applications [http://grouper.ieee.org]. The goal of DSRC standard is to provide wireless communications capabilities for transportation systems within a 1000 meter range at typical highway speeds. VANET have some important characteristics such as nodes forming the networks are vehicles, restricted vehicle movements on the road, high mobility of vehicles and rapid changes in topology and time-varying vehicle density. Since the network topology in the VANETs is frequently changing, finding and maintaining routes is very challenging in VANET. To facilitate communication within a network, a routing protocol is used to find reliable and efficient routes between nodes so that message delivered between them in timely manner. Routing is responsible for selecting and maintaining routes and forwarding packets along the selected routes. Routes

ISSN : 0975-5462

Vol. 3 No. 1 Jan 2011

435

Ram Shringar Raw et al. / International Journal of Engineering Science and Technology (IJEST) between source and destination node may contain multiple hops, this condition is more complex compare to the one hop communication. Intermediate vehicles (nodes) can be used as routers to determine the optimal path along the way.

Fig. 1. Vehicular Ad hoc Networks Scenario [Shaikh Juaid M]

Since the vehicular network topology is frequently and dynamically changing, finding and maintaining routes is very challenging task in VANET. Traditional topology-based routing protocols [Jayakumar (2007)] are not suitable for VANETs. Position-based routing protocols such as GPSR, GPCR, GSR, A-STAR, CAR, MFR, Greedy Routing, etc. are more suitable than other routing protocols. In the past few years many researchers proposed variety of routing protocols. In this paper, we present the significant role of routing protocol, especially the position based routing protocol for vehicular ad hoc networks. The rest of this paper is organized as follows. In section 2, we discuss the basics of V2V communications. Section 3 describes the routing protocols used in V2V communications. In Section 4, we present a comprehensive study of five main position-based routing protocols for V2V communications. Section 5 presents the simulation results. In section 6, we described the comparison and analysis of the position based routing protocols. Finally section 7 concludes this paper. 2. Vehicle-to-Vehicle (V2V) communication Installing fixed infrastructure on roads incurs huge expenses, so V2V communication will be required to extend the effective range of networked vehicles. V2V communication [Zeadally (2010)] is the pure ad hoc communication. This type of communication is mainly used in safety applications like safety warning, traffic information, road obstacle warning, intersection collision warning etc. In V2V communication each vehicle is equipped with GPS (Global Positioning System), sensors, networking devices, digital map which has the road segment information, and computing devices. Vehicles sense its own traffic messages and communicate with its neighboring vehicles by broadcasting beacon or HELLO messages periodically. V2V communication uses both unicast and multi-cast packet forwarding techniques between source and destination vehicles. Unicast forwarding means that a vehicle can only send/ receive packet to/from its direct neighbors. While multi-cast forwarding enables the exchange of packet with remote vehicles using the intermediate vehicles as relays. In V2V communication (see fig. 2), both types of forwarding are used for different type of applications and protocols. The IEEE 802.11p standard is used for V2V communications in highly mobile vehicular traffic environments. Installing fixed infrastructure like access points, base stations, Internet gateways, etc. on roads acquire great expense, so V2V communication will be necessary to extend the effective range of networked vehicles.

ISSN : 0975-5462

Vol. 3 No. 1 Jan 2011

436


Fig. 2. VANET: V2V Communications

Routing Protocols

Topology Based Routing Protocols

Position Based Routing Protocols

Fig. 3. Classification of Routing Protocols

3. Routing protocols in V2V communication Since VANETs change their network topology frequently without any prior information, routing in such dynamic networks is a challenging task. Routing protocols can be broadly classified into two categories: Topology based and Position based routing protocols (see fig.3). 3.1. Topology based routing protocols Topology based routing protocols depend on the information about existing links in the network and use them to perform packet forwarding. The topology based routing protocols can be further subdivided into proactive, reactive, and hybrid protocols. Proactive (table-driven) routing protocols are similar to the connectionless schemes of traditional datagram networks. These protocols employ classical routing strategies such as distance-vector (e.g. DSDV) or link-state (e.g. OLSR) routing and any changes in the link connections are updated periodically throughout the network. Proactive protocols maintain routing information about the available paths in the network even if these paths are not currently used. The main disadvantage of these protocols is the maintenance of unused paths may occupy an important part of the available bandwidth if the network topology changes frequently. However, proactive protocols may not always be suitable for highly mobile networks such as VANETs. Reactive (on-demand) routing protocols (e.g. AODV, DSR) employ a lazy approach whereby mobile nodes only discover routes to destinations on-demand. These protocols maintain only the routes that are currently in use, thus reducing the burden on the network when only a few of all available routes is in use at any time. Reactive protocols often consume less bandwidth than proactive protocols, but the delay in determining a route can be substantially large. In reactive protocols, since routes are only maintained while in use, it is typically required to perform a route discovery process before packets can be exchanged between nodes. Therefore, this leads to a delay for the first packet to be transmitted. Another disadvantage is that, although route maintenance is limited to the routes currently in use, it may still generate a significant amount of network traffic when the network topology changes frequently. Finally, packets transmitted to the destination are likely to be lost if the route to the destination changes. Hybrid routing protocol (ZRP) combines both proactive and reactive approaches to achieve a higher level of efficiency and scalability. However, even a combination of both approaches still needs to maintain at least those network routes that are currently in use. Therefore, limiting the amount of topological changes, that can be tolerated within a given amount of time. However, VANET differ from other networks by its highly dynamic topology. Many simulation result showed that most of the topology based routing protocols suffer from highly dynamic nature of vehicular node mobility because they tend to have poor route convergence and low communication throughput. Position based routing protocols has been identified as a more suitable routing protocols for VANETs to give better performance and exhibit scalability and robustness against frequent topological changes.

ISSN : 0975-5462

Vol. 3 No. 1 Jan 2011

437


3.2. Position based routing protocols Position is one of the most important data for vehicles. In VANET each vehicle wishes to know its own position as well as its neighbor vehicle’s position. A routing protocol using position information in known as the position based routing protocol. Position based routing protocols [Li (2007), Qabajeh (2009)] need the information about the physical location of participating vehicles be available. This position can be obtained by periodically transmitted control messages or beacons to the direct neighbors. A sender can request the position of a receiver by means of a location service. Position based routing protocols are more suitable for VANETs since the vehicular nodes are known to move along established paths. Since routing tables are not used in these protocols therefore no overhead is incurred when tracing a route.

Beacon Control Message

Vehicle ID

Speed

Direction

Location

Current Time

Fig. 4. Major Element of Beacon Control Message

In VANETs, route is composed of several pair of vehicles (communication links) connected to each other from the source vehicle to the destination vehicle. If we know the current information of vehicles involved in the routes, we can predict their positions in the near future to predict the link between each pair of vehicles in the path. VANET is a self-organizing mobile ad hoc network in which to acquire the position information of neighboring nodes, each node periodically exchanges a list of all neighbors it can reach in one hop, using a HELLO control message or a beacon that contains its ID, location, speed, and a timestamp (see fig. 4). One of the main advantages of using position based routing protocol is that it's characteristic of not requiring maintenance of routes, which is very appropriate for highly dynamic networks such as VANETs. Table 1. Comparison of Routing Protocols

Topology Based Routing Protocols Need of route maintenance for all routes. Require large bandwidth if network topology changes.

Position Based Routing Protocols No need of route maintenance.

Forwarding decision is based on the source node.

Forwarding decision is based on the position of destination and the next hop neighbor.

Based on route discovery scheme.

Based on location service scheme.

DSDV, OLSR, AODV, DSR, TORA, ZRP, etc.

GPSR, A-STAR, AMAR, GyTAR, EBGR, MFR, B-MFR, etc.

Does not require large bandwidth.

4. Position based routing protocols for V2V communications Recently, some position based routing protocols such as Greedy Perimeter Stateless Routing (GPSR), Adaptive Movement Aware Routing (AMAR), Improved Greedy Traffic Aware Routing (GyTAR), Edge node Based Greedy Routing (EBGR) and Border node based Most Forward within Radius (B-MFR) routing specific to V2V communications have been proposed.

ISSN : 0975-5462

Vol. 3 No. 1 Jan 2011

438


4.1. Greedy perimeter stateless routing protocol (GPSR) Position-based routing protocols for VANETs highly depend on the knowledge of the neighbor’s positions. This information is updated periodically via HELLO or beacon messages. In GPSR [Karp (2002), Raw (2010a)] (see fig. 5) a node finds the location of its neighbors by means of their HELLO messages and the position of the destination with the help of location service. GPSR requires that each node in the network is able to find its current position by using GPS receiver which provides current location, speed, current time and direction of the vehicles. With all these information, a node forwards incoming packets to a neighboring node closest to the destination, located in a geographical region. This operating mode is known as Greedy Forwarding in which the neighbor which is closest to the destination is selected as the next-hop node. In some cases, when HELLO messages get lost due to temporary transmission errors, some vehicles become unaware of existence of its neighbors. However in some regions of the network, a local maximum may occur when a forwarding node has no neighbor which is closer to the destination than itself. In this situation GPSR uses a most advance recovery strategy called perimeter routing which uses an algorithm of planer graph traversal to find a way out of the local maximum region. Although this advancement, considering only position information may lead packets to be forwarded in a wrong direction and looses therefore, good candidates that ensure its delivery. Since the topology of a vehicular network in urban or city environment is likely to meet local maximum, we have turned recovery strategy of perimeter routing on during our experiments.

Fig. 5. Greedy Forwarding (A is S’s neighbor closest to D).

4.2. Adaptive movement aware routing protocol (AMAR) In the greedy routing scheme, a packet is forwarded to the next-hop neighbor node by unicast manner. In this method a sender node finds the position information of neighbor nodes, and selects the neighbor node which is closest to the destination node as the next hop node. AMAR [Brahmi (2009)] is a Movement Aware Greedy Forwarding (MAGF) based on the greedy forwarding scheme to select next-hop node towards the destination. AMAR scheme makes use of additional information about vehicle movement to select an appropriate packet’s next-hop that ensures the data delivery. This scheme is suitable for highly mobile vehicular ad hoc network and even it performs better in case of pure greedy forwarding failure. In AMAR every vehicle calculates its position, speed and direction by using the GPS or navigation system. Then after its significant role is to assign priority between neighbors while selecting a next-hop node for forwarding a packet. The basic idea of this approach is to compute a weighted score Wi which depends on three factors: the position, the speed, and the direction of vehicle nodes. This weighted score Wi can be computed by current packet forwarder for neighbor node I as follows: Wi = αPm + βDm + γSm

(1)

Where α, β, and γ are the weight of the three used metrics Pm, Dm, and Sm representing respectively the position, the direction and the speed factors with α + β + γ = 1. The AMAR movement aware greedy forwarding improves the data delivery and exploits the concepts of link lifetime to address the inaccuracy of traditional position-based routing and also to avoid sending data to an old neighbor becoming out of the neighbor’s communication range. 4.3. Improved greedy traffic aware routing protocol (GyTAR) Improved GyTAR [Jebri (2006)] is an intersection-based position based routing protocol capable to find robust routes for V2V communications within city traffic environments. GyTAR is based on anchor-based routing scheme with street awareness. GyTAR protocol utilizes two methods for packet transmission: (i) Intersection or

ISSN : 0975-5462

Vol. 3 No. 1 Jan 2011

439

Ram Shringar Raw et al. / International Journal of Engineering Science and Technology (IJEST) Junction Selection: In this method, GyTAR uses junction through which a packet must pass to reach its destination. (ii) Improved Greedy Forwarding Method: Once the destination junction determined, the improved greedy forwarding is used to forward packets between two junctions. GyTAR uses real time traffic density and movement prediction information to forward packet to the destination in VANETs through V2V communications. Hence GyTAR protocol can be used to forward packet successfully to the destination along streets where there are large number of vehicles to provide connectivity. 4.4. Edge node based greedy routing protocol (EBGR) EBGR [Prasanth (2009)] is the position based routing protocol based on greedy forwarding strategy. EBGR protocol uses unicast for sending message from any node to any other node or broadcast for sending message from one node to all other nodes in highly dynamic networks. This method selects the edge node of the limited transmission range as a next hop node for sending message from source to destination. In this method, a packet is sent to the edge node with consideration of nodes moving in the direction of the destination. During packet transmission from source to destination, EBGR uses three basic methods: (i) Neighbor node selection method (ii) Node direction identification method and (iii) Edge node selection method. First method is responsible to collect information of all direct neighbors within the transmission range of the source node. Second method is responsible to identify the direction of moving nodes towards the direction of destination. Finally, third method is used to select edge node as a next hop node within the transmission range for further forwarding the packet. EBGR can be used to minimize number of hops between source and destination and maximizing the network throughput.

B A B’ A’ S

D

Fig. 6. B-MFR Forwarding Method

4.5. Border-node based most forward within radius routing protocol (B-MFR) Next-hop forwarding method like greedy forwarding scheme for linear network does not support well in highly mobile ad hoc network such as vehicular ad hoc network. Therefore, other position based protocols such as MFR, GEDIR, Compass routing, etc. have been used for VANET to improve its performance for non-linear network in a high vehicular density environment. These protocols can be further improvement by utilizing farthest one-hop node in a dense and highly mobile network. Border-node based Most Forward within Radius (B-MFR) [Raw (2010b)] is a position based routing protocol that uses Border-Nodes with maximum projection. The B-MFR utilizes the border-node to avoid using interior nodes within the transmission range for further transmitting the packet. This method selects the border-node as a next-hop node for forwarding packet from source to destination. In this method, a packet is sent to the border-node with the greatest progress as the distance between source and destination projected onto the line drawn from source to destination. In fig. 6, node A is a border-node of source node S, since node A is positioned at maximum transmission range and has maximum progress distance SA’ where A’ is projection of A on SD. Therefore, A is selected as the next-hop forwarding node. Node A is the next-hop forwarding node when it receives the message from S. It uses the same method, to find the next forwarding node with greatest projected distance towards destination. In this case, node B is selected as a border node of A for forwarding packets to destination. Finally node B directly delivers the message to destination node D.

ISSN : 0975-5462

Vol. 3 No. 1 Jan 2011

440


5. Simulation results To evaluate the performance of the above position based routing protocols, it is implemented using NS-2 simulator and simulations are conducted. Here we have taken best three positions based routing protocols (GyTAR, EBGR, B-MFR) out of five for performance comparison in vehicular environment. Based on the simulation parameters given below, we simulate the protocols with a variable transmission range from 100m to 1000m. We consider highway traffic scenario where vehicles are moving on the straight line and a city traffic scenario where vehicles are moving in every direction. Table 2. Simulation Setup

Parameter Simulation time Simulation area No. of Vehicles Vehicle’s Speed Transmission Range No. of Packet Senders Packet Size Vehicle Hello Interval Application MAC Protocol

Values 500s 2500 * 2500 m2 20, 40, 60, 80, 100, 120, 140 40 – 60 Km/h 100m - 1000m 25 512bytes 0.20, 0.40, 0.80 CBR IEEE 802.11, DCF

Fig. 7. End-to-End Delay

The IEEE 802.11 DCF (Distributed Coordinated Function) is used as the MAC protocol. We use a 2500m x 2500m square area for simulation. Network size is represented by the number of vehicles. The speed of vehicles varies from 40-60 km/h. The traffic density is not uniform and it depends on the number of vehicles chosen in the given area. Among all the vehicles, 25 pairs of source-destination are chosen randomly to send packets. The packet transmission density can be adjusted by setting different CBR rates with a packet size of 512 bytes. A simulation runs for 500 seconds and we have taken average of 10 simulation runs. We use two performance metrics to evaluate these three position based routing protocols (GyTAR, EBGR, B-MFR). 5.1 End-to-end delay End-to-end delay is the average delay between source and destination node for all successfully delivered data packets. In fig. 7, the end-to-end delay for B-MFR is significantly lowers than EBGR and GyTAR. Further, BMFR and EBGR has comparatively small end-to-end delay when number of vehicles becomes more. Therefore

ISSN : 0975-5462

Vol. 3 No. 1 Jan 2011

441

Ram Shringar Raw et al. / International Journal of Engineering Science and Technology (IJEST) from this fig. 7, we can observe that B-MFR outperform EBGR and GyTAR both and EBGR outperform GyTAR in terms of end-to-end delay. The improved performance of B-MFR and EBGR in different traffic scenarios can easily be explained by understanding the significance of using border nodes as next hop forwarding node. In case of B-MFR and EBGR border nodes and edge nodes are selected for forwarding packet further that reduces the number of hops in a route. Therefore, in B-MFR and EBGR, the time taken to deliver the packet from source to destination (endto-end delay) is reduced in any traffic scenarios. Further, in B-MFR as the node density increases, the probability of presence of border node increases as compared to EBGR and GyTAR. This gives higher rate of successful deliveries and reduction in number of retransmission. This improves the end-to-end delay that is evident from the fig. 7 as the end-to-end delay for B-MFR grows slowly as the number of nodes increases. 5.2 Packet delivery ratio Packet delivery ratio is the ratio of the packets that successfully reach the destination. Here we compare the performance of GyTAR, EBGR, and B-MFR in terms of packet delivery ratio. From the fig. 8, we can see how packet delivery is affected by the packet transmission density and vehicular traffic density. In case of low vehicle density, very few vehicles will be available within the transmission range for next-hop selection along a particular path. When the vehicle density is more, the connectivity is much better. In this case all routing methods achieves better delivery ratio, since more vehicles can be met to forwards packets. In B-MFR and EBGR, a node will forward packet to the next-hop border node or edge node of its transmission range which is moving towards the destination. The packet delivery ratio is directly proportional to the vehicle density. As shown in fig. 8, B-MFR outperforms EBGR and GyTAR both and EBGR outperform GyTAR in terms of packet delivery ratio when the vehicle density is high.

Fig. 8. Packet Delivery Ratio

6. Comparison and analysis The objective of a routing protocol is to guarantee a reliable and efficient delivery of packets. A routing algorithm can be evaluated based on performance metrics such as hop count, packet delivery ratio, end-to-end delay and overhead packets required. However each routing protocol for VANET has different features and requirements, suited for different vehicular traffic scenarios. For comparison, we have selected some position based routing protocols such as GPSR, AMAR, GyTAR, EBGR and B-MFR out of the many reviewed. Table 3 and 4 summarizes the discussed routing protocols. GPSR uses greedy forwarding with most advanced recovery strategy called perimeter mode. But GPSR using perimeter mode is relatively incompetent in highly dense V2V networks. GPSR has low packet delivery rate and high latency. Also GPSR has limitations in the aspects of large number of hops, wrong direction, routing loops, etc. AMAR is the movement aware greedy forwarding (MAGF), designed to match the highly dynamic network requirements. AMAR outperforms GPSR in terms of packet delivery ratio and end-to-end delay. Improved

ISSN : 0975-5462

Vol. 3 No. 1 Jan 2011

442

Ram Shringar Raw et al. / International Journal of Engineering Science and Technology (IJEST) GyTAR protocol uses real time traffic density information to route data in high dynamic VANETs. Based on the GPS, GyTAR aims to efficiently forward packets in the real time road traffic networks. GyTAR achieves the highest packet delivery ratio for the different nodes in the networks as compared to GPSR and AMAR. B-MFR and EBGR are very well suited for VANET. Both can minimize the number of hops and deliver the packet at low latency. But B-MFR is more efficient than EBGR because it uses only the exact border node to forward the packet from source to destination. Table 3. Comparison of Position Based Routing Protocols in V2V

Position Based Routing

Forwarding Strategy

GPSR

Greedy forwarding Greedy forwarding Improved Greedy Forwarding Greedy forwarding Greedy forwarding

AMAR GyTAR EBGR B-MFR

Position Information

Mobility Model

Network Simulator

Right hand rule

Packet forwarding

Random waypoint

Ns-2

Movement awareness

Packet forwarding

Unknown

Ns-2

Carry and Forward

Packet forwarding

Realistic Mobility Model

Ns-2

End node awareness

Packet forwarding

Manhattan Mobility Model

NCTUns 5.0

Border node Awareness

Packet forwarding

Realistic Mobility Model

NS-2

Recovery Strategy

Table 4. Performances of Position Based Routing Protocols

Metric

Low

Hop Count

B-MFR

End-to-End Delay

B-MFR, EBGR GPSR, AMAR B-MFR, EBGR B-MFR, EBGR

Delivery Rate Overhead Packet Latency

Medium AMAR, GyTAR AMAR, GyTAR GyTAR AMAR GyTAR, AMAR

High GPSR GPSR B-MFR, EBGR GPSR, GyTAR GPSR

7. Conclusion and future works VANETs will play a significant role for automotive industries and in real traffic scenarios. The routing of data packets in VANETs is challenging and subject of intensive research. In this paper we have taken several position-based routing protocols recently proposed for V2V communication among vehicular ad hoc networks (VANETs). We have presented a comparison of five categories of position-based routing protocols, highlighted their result, features, differences, and characteristics. Among these five protocols we have simulated best of three protocols (GyTAR, EBGR, B-MFR). Performance of these three routing protocols has shown in terms of end-to-end delay and packet delivery ratio. Result shows that for comfort and safety applications in a V2V communication environment the significant requirements are high and reliable packet delivery rate at a minimal latency. In general, position-based routing is more promising than other routing protocols for VANETs because of the geographical constrains. These routing protocols would improve the traffic control management and provide the information in timely manner to the concern authority and drivers. From the result and analysis, we now look into further improving the use of position-based routing protocols in the V2V communication of the vehicular environment. These routing protocols ensure that all the packets are received with small delay to stop accidents on the road. The scope of the research on VANET would be a position-based routing protocol that is used for comfort and safety applications.

ISSN : 0975-5462

Vol. 3 No. 1 Jan 2011

443


References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15]

Abdalla G.M.T. Abu-Rgheff M.A. and Senouci S.M.. ( 2008):Current Trends in Vehicular Ad Hoc Networks. Brahmi N., Boussedjra M., Mouzna J., Bayart M ( 2009): Adaptive Movement Aware Routing for Vehicular Ad Hoc Networks Leipzig. Germany. Jayakumar G., Gopinath G. ( 2007): Ad Hoc Mobile Wireless Networks Routing Protocols – A Review. Journal of Computer Science 3 (8): pp. 574-582. Jebri M., Meraihi R., Senouci S., Ghamri-Doudane Y. ( 2006): GyTAR: improved Greedy Traffic Aware Routing Protocol for Vehicular Ad Hoc Networks in City Environments . VANET’ 06, Los Angeles, California, USA. Karp B., Kung H.T. ( 2002): GPSR: Greedy Perimeter Stateless Routing for Wireless Networks. MobiCom. Li . F., Wang.Y. ( 2007): Routing in Vehicular Ad Hoc Networks: A Survey. IEEE Vehicular Magazine. Olariu S., Weigle M. C. ( 2009): Vehicular Networks: From Theory to Practice. CRC Press, A Chapman & Hall Book. Papadimitratos P., Buttyan L., Holczer T., Schoch E. , Freudiger J.,Kargl F. ,Ma Z. , Kung A. and Hubaux J. ( 2008): Secure Vehicular Communication Systems: Design and Architecture. IEEE Communications Magazine. Prasanth K. , Duraiswamy K., and Jayasudha K., K. Handrasekar K. ( 2009): Minimizing End-to-End Delay in Vehicular ad Hoc Network using Edge Node Based Greedy Routing. IEEE. Qabajeh L.K., Kiah L.M., and Qabajeh M.M. ( 2009): A Qualitative Comparison of Position-Based Routing Protocols for Ad-Hoc Networks. IJCSNS. Raw R.S., Lobiyal D.K., ( 2010): VANET: Position-Based Routing in Urban Environment – A Technical Review. TRANSTEC 2010, New Delhi. Raw R.S., Lobiyal D.K. ( 2010):B-MFR Routing Protocol for Vehicular Ad hoc Networks . IEEE, ICNIT 2010, Manila, Philippines. Shaikh Juaid M.: Minor thesis presentation. A Comparative Analysis of routing protocols in VANET environment using realistic vehicular traces.ppt. Task Group p, “IEEE 802.11 p Wireless Access for Vehicular Environment”, Draft Standard, ttp://grouper.ieee.org/groups/802/11/. Zeadally S., Hunt R., Chen Y., Irwin A., and Hassan A.,( 2010): “Vehicular Ad Hoc Networks (VANETS): Status, Results, and Challenges,”.

ISSN : 0975-5462

Vol. 3 No. 1 Jan 2011

444