You are on page 1of 21

1.Introduction to Vehicular Adhoc Network.

Vehicular Ad-Hoc Network (VANET) is a subset of mobile ad-hoc network, which supports data communications among nearby vehicles and between vehicles and nearby fixed infrastructure, and generally represented as roadside entities. Depending on the range of data communications, nodes in VANET communicate among themselves in type of short-range (vehicle-to-vehicle) or medium-range (vehicle-to-roadside) communications.The vehicles can communicate with each other in addition to communicating with available supporting infrastructure along the road. Vehicle-to-vehicle (V2V) communication helps in sharing the information within the VANET, while vehicle-to-infrastructure (V2I) communication extends the VANET to include other designated networks (e.g. a centralized network or the Internet) and allows sharing of information with them..

Fig.1.1:Vehicular Adhoc Networks.

To realize communication in VANETs, the Federal Communications Commission (FCC) dedicated 75MHz of the frequency spectrum in the range 5.850 GHz to 5.925 GHz to be used for V2V and V2I communication. The 5.9 GHz spectrum was termed Dedicated Short Range Communication (DSRC) and uses IEEE 802.11p. VANET applications must share the allocated
1

bandwidth, making it a scarce resource that should be managed very carefully. Inefficient data dissemination wastes a large amount of bandwidth that if saved would allow more vehicular applications to co-exist in addition to allowing the vehicular data to be disseminated further. Many VANET applications require each vehicle to share its data (e.g. speed and location) with its neighbors through broadcasting a message containing such data. Sending these messages to farther distances will waste the bandwidth and may cause a broadcast storm problem based on the traffic density. So, to share the data with vehicles at farther distances efficiently, many data aggregation techniques have been proposed. Data aggregation has been proposed in VANETs to solve the bandwidth utilization problem. The basic idea is to gather information about many vehicles into a single frame. Data aggregation techniques can be classified as syntactic or semantic . Syntactic aggregation uses a technique to compress or encode the data from multiple vehicles in order to fit the data into a single frame. This results in lower overhead than sending each message individually. In semantic aggregation, data from individual vehicles is summarized. For instance, instead of reporting the exact position of five vehicles, only the fact that five vehicles exist is reported. The trade-off is a much smaller message in exchange for a loss of precise data.Efficient data dissemination in VANETs is an important problem that needs to be handled carefully. The importance of this problem has attracted many researchers to study it and, as a consequence, many data dissemination techniques have been proposed. The proposed techniques can be categorized according to the communication method used V2I, V2V, and hybrid of the two. In the infrastructure-based techniques, vehicles mainly communicate through infrastructure units that can be road side units (RSUs) embedded sensor belts in the road pavement , or cellular networks. V2I approaches depend upon a pervasive infrastructure that may not become a reality due to its high cost. V2V techniques for data dissemination depend on data routing or on broadcasting. In the majority of VANET applications (especially safety applications), the exchanged messages have no specific destination but have all the surrounding vehicles as the targeted destinations. This makes broadcast the most suitable method for dissemination 1.1 Application areas of VANET The applications of vehicular ad hoc networks (VANETs) can be classified into three main categories:

2

and road condition reports. vehicle collisions. 1. Some research focuses on Cooperative Collision Avoidance (CCA) to broadcast collision avoidance messages in a very short latency in order to save as many victim vehicles as possible. if it has to depend on the application’s information. A.2 Security Issues in Vehicular Network. multimedia streaming . entertainment. include collision warning and merge assistance. as well as gathering and disseminating weather information. that is to say the packets should be successfully received by the receiver. so it prevents critical information from arriving. lane changing assistance. route planning. c) The third type of applications. It also increases the danger to the driver. The data transmission requirement of this type of service is reliability. For the most urgent situations. such as entertainment.a) General data routing services: The general data routing provides one-to-one data routing or one-to-all data broadcasting for services. For instance. the data should not only be successfully received by the receivers. it can make an accident and use the DoS attack to prevent the warning from reaching to the approaching vehicles. but be received in a very short time to provide the driver with more reaction time.. if a malicious wants to create a massive pile up on the highway.g. which are the main focus of the US DoT’s efforts in vehicular networks. and P2P file sharing These applications are usually life critical.2. See figure 1. 3 . Therefore. the limit of propagation time of the emergency message is extremely low.e. ATTACKS AND THREATS 1) Denial of Service attack This attack happens when the attacker takes control of a vehicle’s resources or jams the communication channel used by the Vehicular Network. b) Safety applications: The safety applications provide one-to-all emergency message broadcasting for receivers in a predefined region. includes Internet access. and communications Informational applications include notification of upcoming traffic conditions and roadway hazards. electronic brake lights. such as. The need for a robust VANET networks is strongly dependent on their security and privacy features.Safety applications.

3) Fabrication Attack. the information could be false or the transmitter could claim that it is somebody else. and use it in another time. For instance. Adversaries 1) Selfish Driver 4 .2:DOS Attack 2) Message Suppression Attack An attacker selectively dropping packets from the network. Basic 802. warnings. or altering the actual entry of the data transmitted 5) Replay Attack This attack happens when an attacker replay the transmission of an earlier information to take advantage of the situation of the message at time of sending.Fig1. This attack includes fabricate messages. these packets may hold critical information for the receiver. replaying earlier transmission. certificates. 4) Alteration Attack This attack happens when attacker alters an existing data. The goal of such an attack would be to confuse the authorities and possibly prevent identification of vehicles in hit-and-run incidents. B. Identities. An attacker can make this attack by transmitting false information into the network. an attacker may suppress a congestion warning. the attacker suppress these packets and can use them again in other time.11 security has no protection against replay. so vehicles will not receive the warning and forced to wait in the traffic. It includes delaying the transmission of the information. The goal of such an attacker would be to prevent registration and insurance authorities from learning about collisions involving his vehicle and/or to avoid delivering collision reports to roadside access points.

but still the mobility is limited. and maybe will not happen again.1. so they must choose an alternate route.3: Selfish Driver 2) Malicious Attacker This kind of attacker tries to cause damage via the applications available on the vehicular network. and these two vehicles may never meet again. regardless the cost for the system by taking advantage of the network resources illegally. See figure1.The general idea for trust in Vehicular Network is that all vehicles must be trusted initially. and can move from one place to another within the coverage area. these attackers will have specific targets. vehicles make connection throw their way with another vehicles that maybe never faced before. in Vehicular Ad Hoc Networks nodes moving in high mobility. some drivers try to maximize their profit from the network. and maybe will travel in opposite 5 . 2) Volatility The connectivity among nodes can be highly ephemeral. these vehicles are trusted to follow the protocols specified by the application. Vehicles traveling throw coverage area and making connection with other vehicles. In many cases. A Selfish Driver can tell other vehicles that there is congestion in the road. these connections will be lost as each car has a high mobility. and they will have access to the resources of the network. and this connection lasts for only few seconds as each vehicle goes in its direction.3 Fig. So securing mobility challenge is hard problem. 1. so the road will be clear for it.3 Vehicular Networks Challenges 1) Mobility The basic idea from Ad Hoc Networks is that each node in the network is mobile.

another problem arise when we must know that there is no a global authority govern the standards for this network . where the application layer is unreliable. and this will be impractical for securing VC. causes an overhead. the receiver verifies the message. 4) Bootstrap At this moment only few number of cars will be have the equipment required for the DSRC radios. for many applications Vehicular Networks will require realtime. to do this vehicles will assign every message with their Private Key along with its Certificate. and this number is growing. to get a financial benefit that will courage the commercial firms to invest in this technology 1. in the future we must concentrate on getting the number higher. so personal contact of users device to a hot spot will require long life password. the problem is much bigger. a delay in millisecond makes the message meaningless. In some messages. to reduce this overhead we can use the approach ECC (Elliptic Curve Cryptography). for example: the standards for DSRC in North America is deferent from the DSRC standards in Europe.direction. a delay in seconds for some applications will make the message meaningless and maybe the result will be devastating. Vehicular networks lacks the relatively long life context. 2.4 Security Requirements 1. Signing each message with this. the standards for the GM Vehicles is deferent from the BMW one. so if we make a communication we have to assume that there is a limited number of cars that will receive the communication. these applications need faster response from Sensor Networks or even Ad Hoc Network. since the 6 . at the receiver side. to make sure for its origin and to control authorization level of the vehicles. the vehicle will receive the message and check for the key and certificate once this is done. the efficient public key cryptosystem. Authentication In VC every message must be authenticated. Availability Vehicular Network must be available all the time. Attempting to meet real-time demands makes the system vulnerable to the DoS attack. 2) Network Scalability The scale of this network in the world approximately exceeding the 750 million nodes.

trip path. thus multi-hop broadcasting is required during the communication. to identify the vehicle in anywhere . time. it 7 . this license is installed in the factory for every new vehicle. an emergency message (or messages) will be generated by an observer vehicle and be broadcast. Confidentiality The privacy of each driver must be protected.For preserving the real identity of the driver. this information like real identity. 3. speed etc. 2. speed. As the observer detects an automobile accident occurring.1 shows an emergency message forwarding scheme.The privacy could be achieved by using temporary (anonymous) keys. 4. an ELP (Electronic License Plate) is used. with the RFID technology to hold the ELP.Emergency Message Forwarding Schemes in VANET.all the keys will be stored in the TPD. called the risk zone (usually several kilometers). Figure 2. these keys will be changed frequently as each key could be used just for one time and expires after usage . Non-repudiation Non-repudiation will facilitate the ability to identify the attackers even after the attack happens. Privacy Keeping the information of the drivers away from unauthorized observers. Since the message might not be relevant to all areas. For safety applications emergency messages are broadcasted. Generally the one-hop broadcast range is just several hundreds of meters and cannot cover the entire risk zone. and will be reloaded again in next time that the vehicle makes an official checkup . Any information related to the car like: the trip rout.This prevents cheaters from denying their crimes. the messages should be encrypted to prevent outsiders from gaining the drivers information.potential way to recover with unreliable transmission is to store partial messages in hopes to be completed in next transmission. When an emergency event occurs. any official side holding authorization can retrieve this data. it will provide an identification number for the vehicle. any violation will be stored in the TPD. 5. the emergency message is only broadcast within a limited region.

Emergency message forwarding scheme for safety application in VANET Similarly. the additional waiting time before rebroadcasting the message should also be as small as possible to ensure low latency. Moreover. In order to ensure all the vehicles in the risk zone can receive the emergency message in time. we consider two important ways to reduce the message propagation latency. However. A large number of receiving nodes broadcasting the same message will create the problems of channel contention and serious packet collisions. The above tasks continue until the message reaches the edge of the risk zone. the new chosen forwarder will broadcast the message and also select the next forwarder for message relaying. without appropriate forwarder node selection. Since the involved vehicles of message broadcasting are only the forwarders. makes the problem even worse.immediately broadcasts an emergency message to inform all its neighbors. FORWARDING SCHEMES 8 . Rebroadcasting due to transmission failure.1. The key point is to reduce the hop counts needed to propagate the message to the entire risk zone. Fig2.The scheme vehicle-density-based emergency broadcast scheme called the VDEB is proposed to reduce the message propagation latency with little overhead. This can be done by choosing the forwarders such that the hop distance is maximized. the broadcast storm problem can then be relieved. there will be many nodes rebroadcasting the message. meanwhile one of the observer’s neighbors (the forwarder) will be selected to play the role of message relaying.

Among the papers concerning the sender-oriented scheme. They calculate a waiting time.1 The sender-oriented schemes Sender-oriented schemes use accurate neighbor positions to select the best forwarder node (usually the farthest node from the sender) claiming minimum hop count and no additional waiting time. They use periodic beacon messages to acquire the vehicle ID of the farthest vehicle in the transmission range of a vehicle. The node with the shortest waiting time may rebroadcast the message first. several vehicles may 9 . Note that the vehicle with the larger waiting time will not necessarily rebroadcast the message first. The papers are MCDS-like (minimum connected dominating set like) methods. in a sparse network. The vehicle nodes located in the backbone will be responsible for message broadcasting. Contrasting with the sender-oriented schemes. Due to the requirement of frequent beacon message exchanges. number of nodes. and try to maintain an MCDS structure as the backbone. this type of method encounters the problem of high overhead. and wait this time period before rebroadcasting the message. Statistically. The receiver-oriented schemes The receiver-oriented schemes use contention to automatically elect the forwarder(s) in a distributed fashion. and so on. All the one hop receivers concerning the emergency event enter the contention phase after receiving the message in the receiver oriented schemes. B. vehicle velocity. The parameters for waiting time calculation can be the distance between the sender and receiver.Several research papers have proposed methods for emergency message forwarding. vehicle’s moving direction.These methods can be categorized into two types: sender-oriented schemes and receiver oriented schemes 2. However. They try to create and maintain stable clusters (or virtual backbones) to reduce the overhead of the backbone structure maintenance. The emergency messages are rebroadcast by the farthest vehicle of each cluster head. This approach uses high frequency beacon messages or handshaking mechanisms to choose a single node as the forwarder. are backbone-based methods. the furthest node from the sender will have the highest chance of rebroadcasting the emergency message first. Moreover. the sender does not assign the forwarder in the broadcast message. the waiting time still has a high chance of being long. All the other nodes overhearing this broadcast will cancel their broadcasting process.

the vehicles should also be equipped with several sensors for vehicle velocity. 10 . The aim of the ring-based approach is to partition the transmission range of the current forwarder into multiple concentric rings. it synthesizes the main ideas of the sender-oriented scheme and tries to reduce the influence of the drawbacks of both methods. and the overhead will therefore become large. • • • Assume all of the vehicles on the road are equipped with a positioning device such as GPS to acquire their own positions.have a similar contention window size. For the transmission of packets. This might cause network congestion and collisions. On the other hand. however. The Vehicle Density Based Emergency Broadcasting Scheme(VDEB) The considered scenario of our method is in a highway environment. The sender-oriented schemes mainly rely on accurate position data to select the forwarder. The forwarders in a sparse network have a higher chance of being far from the border of the sender’s transmission range because the inter-vehicle spaces are larger than in a dense network. Both types of the existing message forwarding schemes have their drawbacks. etc. the VDEB adopts a ring-based approach to shorten the waiting time. 3. the latency of the receiver-oriented schemes usually gets much longer in a sparse network. To detect an emergency event. TheVDEB is a receiver-oriented forwarding method. the situation of multiple rebroadcasting also could occur. The waiting time is usually reversely proportional to the distance between the sender and the receiver. inaccurate data may cause the sender to select a forwarder outside the transmission range and thus the forwarding process will fail. In addition. As the vehicle density increases.. the broadcasting period of the beacon messages should be very short. The basic idea is that. To maintain accurate data. each vehicle is equipped with the WAVE DSRC device. in order to overcome the problem of a long waiting time in a sparse network for the receiver-oriented scheme.

1: The ring-based approach of the VDEB. not to keep track of each vehicle’s correct position.1. we adopt a vehicle position prediction mechanism in the VDEB forwarding scheme. the forwarder has to maintain a neighbor table. The length of the waiting time is increased for each ring number. Thus the frequency of the beacon message exchange for maintaining the neighbor table is less compared with in the traditional sender-oriented scheme. say ring 0. have the shortest waiting time. the vehicles in ring number greater than and equal to i + 1 will not rebroadcast the message.For the example in Figure3.3. if there are vehicles in ring i.Fig. 11 . then their messages would collide. Note that. it will be broadcast to notify the neighbors of the forwarder. the transmission range is partitioned into 5 rings. If any vehicle successfully broadcasts the message. The gap between two adjacent rings is called the ring width. As the ring width is determined. they can use an 802. Usually. The computation of the ring width is performed at the forwarder. The receiver can then compute the waiting time according to the received ring width. if two (or more) vehicles in the same ring come out. If no successful transmissions are done by the vehicles in ring 0. The vehicles in the outermost ring. Each ring is assigned a timeslot. In order to obtain the ring width. The main objective of maintaining the neighbor information is to determine the vehicle density in the transmission range of the current forwarder. the vehicles in ring 1 will forward the message. In this case.11 like backoff mechanism to resolve their collisions. other vehicles will cancel their waiting process since the forwarding is done without collisions. Besides.

With the position information. every vehicle can roughly estimate the number of neighboring vehicles in its transmission range. y0) + (t − t0) × (vx. The validity of the neighbor information is verified by checking it in the neighbor table to see if its lifetime has expired or not. an emergency message has the information about the event. The final field is the time stamp t0. then it will not be a neighboring vehicle. velocity. The third field is the vehicle’s velocity (vx. (x. vehicles only care about the presence of neighboring vehicles. the receivers can roughly estimate the current position.2 The emergency message format and the ring formation Generally.2 The hello message format. The accuracy of the position information is reached by frequently transmitting the hello message. such as the position of the event. we consider that the emergency message 12 .3. If the current position of a vehicle is not found in its transmission range. Then it can be used to count the number of neighboring vehicles around one vehicle (called vehicle density in this paper). and the risk zone (RZ) size of the event. To reflect the characteristics of an emergency event. y0). The first and the second fields of the hellomessage denote the vehicle’s identity and its position (x0.1 Maintenance of the neighbor table In this approach. In other words. and the time stamp. the event emergency level. In this type of scheme. y) = (x0. if the same vehicle information is received.2. the event description. Fig. The hello messages in sender-oriented schemes usually have position information and the timestamp but not velocity. 3. otherwise the entry for this vehicle will just be removed. The beacon messages are called hello messages in based on their functionality. By employing the velocity of the neighboring vehicle to estimate its position after a period of time. vy) when the message is generated. The format of the hello message is given in Figure 3. 3. vy) ………(1) where t is the current time.respectively.the sender using the following equation. the purpose of the beacon message is to inform other neighboring vehicles to roughly estimate the future position of vehicles in their neighbor table. the position information and the timestamp will be updated.

4. The maximum ring width appears in the case that all the vehicles drive side by side (see Figure 3. we can easily determine the maximum and minimum ring width as follows. Consider the same case of inter-vehicl distance.3. 3. The ring width is included in the message for receivers to calculate their waiting time. We assume that the vehicles have identical inter-vehicle distance with its immediate front and back vehicles.3:Emergency message format.4). the position of the current forwarder. the Fig.3. and N is the number of vehicles in the transmission range. we also add the IDs of the source and the current forwarder. Under this assumption. l is the number of lanes. Fig. Besides the above information. Hence the maximum ring width can be calculated by: MaxRingWidth = R×l/N where R is the transmission range radius. and the ring width to the emergency message. The position of the current forwarder for receivers is used to determine in which ring they are located. The possible cases of the maximum and minimum ring widths. 13 .should be rebroadcast until it reaches the edge of the risk zone. The detailed emergency message format is shown in Figure 3.

3 The emergency message broadcasting scheme The VDEB uses hello messages to periodically exchange the basic information between any two vehicles. The waiting time of the other rings is proportional to their ring number as follows: WaitingT ime = SlotTime × RingNumber …. } Compute the vehicle density N. As the next forwarder is elected. /* Determine the ring width value */ MaxRingWidth = R×l/N 14 .(4) When a vehicle waiting in the waiting process overhears any vehicle forwarding an emergency message. the message relaying will be continued until the border of the risk zone is reached.MaxRingWidth] and then the emergency message can be broadcast to the neighbors of the current forwarder. The vehicles in the second and third lanes partition the inter-vehicle distance of the first lane into three equal parts. The waiting time of ring 0 is 0 SIFS time.4. Upon receiving an emergency message from a vehicle. 3. For the current forwarder. The purpose of simulations is to compare the performance of our proposed VDEB with the other conventional emergency message forwarding schemes. The minimum ring width can be calculated by: MinRingWidth=R/N The ring width can be randomly set to be the value in [MinRingWidth. the vehicle density can be easily obtained from its neighbor table and then the ring width can be determined.minimum ring width appears when the vehicle distribution is as shown in Figure 3. each vehicle can maintain its own neighbor table. it will be added to the emergency message and then be broadcast to the one-hop neighbors. it firstly calculates its ring number by the ring width and the position of the forwarder given in the message. { if (the current vehicle is the observer) { Add event-related information into the EM. the waiting process is canceled. As the ring width is estimated. Based on the information in the received hello message. A) Procedure for forwarder-emergency-message-forwarding Output: emergency message (EM).

2s to 25. Simulation Environmental Setup: The simulation scenario is an 8km. We vary the hello interval from 0. 3 lane. 4. { Retrieve RingWidth value from the EM. The mobility of vehicles is evaluated by two speed scenarios: the low speed scenario has a speed of [20km/h. The risk zone of an emergency event is 1km. Perform the Waiting process. single direction section of highway. call procedure forwarder-emergency-message-forwarding. Since the performance results for both scenarios (the low speed and the high speed) are 15 . Input: emergency message (EM).2s for the MCDS scheme. RingNumber=Ring-Number-Determination(RingWidth. l is the number of lanes. Add RingWidth into the EM.6s . while the high speed scenario has a speed of [70km/h. current forwarder’s position).MinRingWidth = R N . The vehicle density varies from a sparse to a dense road environment based on the safety distance with respect to the vehicles’ speed. 120km/h]. 70km/h]. Broadcast(EM). The VDEB is also simulated with the hello interval from 3.MaxRingWidth].1. } B) Procedure for receiver-emergency-message-forwarding. WaitingTime=SlotTime × RingNumber. } else cancel the waiting process.Simulation Analysis 4. RingWidth= randomly depicts any value in [MinRingWidth. if ((the waiting timer expired) and (does not overhear EM from neighbor)) { Set itself to be the forwarder. where R is the transmission range.2s to 3.

2. the impact of varying the Table 1: List for the simulation parameter settings.2. 130] vehicles/km [20. 70]. 120] km/h CWmin (32 SIFS) 16 .8s.4s. As shown in Figure 4. 25. 4. Simulation scenario Simulation area Risk zone size Transmission range MCDS hello interval VDEB hello interval Slot time Simulation time Vehicle density Vehicle speed Congestion window highway 8km × 3 lanes 1km 250m 0. thus we only demonstrate the results of the high speed scenario. 0. As shown in Figure 8.2s. [70.2.2s 3. Similarly.1 shows the average delay for the MCDS scheme as the hello interval (HI) varies. 0. a shorter hello interval will disseminate many hello messages in the network.4 is a proper choice.1 The simulations for hello interval setting: Figure 4. 12. which will cause the maintenance to be too costly. the lower average delay we have.2s. the performance of the MCDS when the hello interval is set to 0.2 gives the average delay of the VDEB scheme as the hello message varies. since the average delay is under 4ms for any vehicle density.8s. 3. 1.quite similar. The complete simulation setting is summarized in Table 1.6s 64 SIFS 150s [20. 6. Figure 4.2. Clearly. However. the shorter the hello interval we set.6s.2 The numerical results 4.4s.

Fig. Fig.2. 17 .4.3 and 4. The farthest delay is defined as the average of the delay time in which the emergency message reaches the farthest receiver. we can use larger hello intervals in our VDEB scheme. VDEB.1: The average delay comparisons of the different hello intervals of MCDS.which will result in less network overhead consumption.2 The simulation results for performance metrics comparison: Figures 4. The hello interval of VDEB will therefore be set to 6.2.2: The average delay comparisons of the different hello intervals of VDEB.2. To achieve a similar delay performance to the MCDS scheme. and DBS. 4.2.4 for the second part of the simulation.2.4 show the average delay and the farthest delay simulation results for the MCDS.4.

Fig. we learnt that the average delay and the farthest delay of DBS is the worst compared to the MCDS and VDEB schemes.3 and 4.5 gives the results of the average retransmission comparison. our proposed VDEB scheme outperforms the other schemes. From Figures 4.4: Performance comparisons of the farthest delay Figure 4.2.4 .2.2. In addition.2.4. Fig. 18 .4.3: Performance comparisons of the average delay.2.

Thus the VDEB consumes very 19 . so the multipleforwarder problem will occur. Our VDEB scheme is a receiver-oriented method. the number of retransmissions of DBS increases when the vehicle density gets higher.4.4 seconds). 000. it will not cause multiple forwarders. Fig. the number of total transmitted hello messages in MCDS (hello interval is 0.6: Performance comparisons of the overhead. because the next forwarder is chosen before broadcasting the message. there are only 24. In the MCDS scheme. This is because there are more vehicles with a similar waiting time if the vehicle density gets higher. while in VDEB (hello interval is 6. the number of average retransmissions is not greater than 15 for both the MCDS and VDEB schemes. Regardless of the vehicle density. The comparison results for the number of total transmitted hello messages are summarized in Figure 13.5: Performance comparisons of the average retransmissions The DBS also performs poorly in the number of retransmissions. 375 hello messages transmitted. As shown in Figure 4.2.6.4.2. When the vehicle density is 130 vehicles per kilometer.4 seconds) is 390.2.Fig.

Conclusion The sender-oriented schemes have the drawback of high overhead. and the receiver oriented schemes cause longer delays. 5. VDEB scheme resolves both of these problems and provides lower 20 .low overhead to obtain even better performance than MCDS at the expense of a few more retransmissions.

the number of forwarders is limited by the ring width which is estimated by the vehicle density and neighbor. In addition. Vehicle density help to reduce the number of retransmissions of messages. In this a receiver oriented contention mechanism is adopted with the vehicle density measurement component.delay and lower overhead for emergency message broadcasting. and avoid the situation that the delay time increases if the forwarder is not far enough away. In the VDEB scheme .the number of retransmission is not proportional to the vehicle density information 21 .