Research On Trusted Routing Protocols In VANETs

Currently, the road traffic accidents have become a global public security issue, and vehicle ad hoc networks called VANETs have emerged as an important part of Intelligent Transportation Systems. Compared with the traditional Internet, VANETs have many different characteristics, such as the unstable wireless channel quality, vehicles distribution limited by roads, nodes moving fast, frequent changes in the network topology, short path lifetime, broadcasting packets blocked by flood transmission, abundant external auxiliary information, no energy constraints and so on. So the routing problems in VANETs are facing important challenges.There are many problems in establishing a reliable and trusted routing model in VANETs. Firstly, the routing information is vulnerable because of the frequently changing topology. Malicious vehicles disseminate untrue information and disrupt routing information. Selfish vehicles refuse to cooperate with other vehicles and terminate routing forwarding behaviors. These behaviors may cause network interruption and loss of data. Because VANETs have the self-organizing and no-centers characteristics, the existing methods based on centralized solution is no longer applicable. Secondly, single hop inter-vehicle transmission usually can not cover all destination nodes because of the limited vehicle transmission scope. So it is one of the basic tasks in VANETs that how to establish multiple hops reliable broadcast model. However, the high speed of vehicles causes the network topology changing frequently, and aggravates channel conflicts because of the flooding broadcast. And serious conflicts generated by multiple hops transmission cause enormous loss of data. Meanwhile, because inter-vehicle communication is confined by road layout constrains, the information is hard to across the different sections of roads in VANETs. Based on the above problems, this paper proposes two novel models in VANETs, one is the trusted routing model based on experienced trustiness for resisting selfish nodes effectively, and the other is the reliable broadcast routing model based on gain prediction for reducing effectively channel conflicts and information unnecessary retransmission. The two above models provide secure and reliable guarantees for inter-vehicle communication.The following describes the work of this paper in detail:1. This paper introduces the idea of the experienced trustiness as the reliability standards, to guide nodes to select the nodes with the high reputation to join in the routing forwarding process. The introduction of the idea of experienced trustiness considers the following factors, (1) the survival time of vehicles, (2) the ratio of the successful interaction among vehicles, and (3) the reputation of the last hop. Thus the well-behaved nodes will improve the value of the experienced trustiness, and obtain the higher reputation due to the contribution to VANETs. On the contrary, the selfish behaviors will lead to reduce the experience trustiness value along with the growing survival time.2. This paper proposes the trusted routing model based on the experienced trustiness called E-GPSR model in VANETs. At present, the most widely used routing protocol is the position-based routing called PBR, and Greedy Perimeter Stateless Routing called GPSR is a representative of PBR. This paper uses the idea of the experienced trustiness in GPSR protocol, and proposes a novel route forwarding method to encourage cooperative behaviors and to discipline selfish behaviors.3. In order to reduce the unnecessary retransmission of data and the probability of channel conflict, this paper presents the idea of the gain prediction to guide nodes to establish the reliable broadcasting model in VANETs. The idea of the gain prediction reflects the coverage scope of the nodes the next time, in the premise that neighbor nodes can receive the broadcast packets from their last hops. It is composed of the direct gain and the indirect gain. The gain prediction focused on three factors:the relative speed, the relative distance between the forwarding nodes and their neighbors, and the difference between the set of covered nodes.4. This paper proposes a reliable broadcast routing model based on the gain prediction called RB-GP model from the reliability aspects of inter-vehicle communication and the coverage scope aspects of forwarding nodes. RB-GP model selects the largest gain nodes in all directions as the rebroadcast nodes to ensure the largest coverage scope during broadcast processes and the lowest delay time of broadcast packets.5. Based on the above preparations, this paper adopts NS2 as the simulation platform and VanetMobiSim as the traffic simulation software to simulate the city, highway and urban scenarios, and analyzes the performances of E-GPSR model and RB-GP model. This paper analyzes the data extracted from the simulation results and compares the packet delivery ratio, the packet forwarding count, the routing packet transmission ration, the average latency and the end-to-end throughput with different scenarios of inter-vehicle communication. Compared with the original model we can conclude that:(1) E-GPSR can identify selfish nodes and establish trusted routes among strangers. (2) E-GPSR model has significantly improved the network performance. (3) The network performance of RB-GP model is better than the reliable broadcast routing model based on the mobility prediction called RB-MP model and Flooding model. To a certain degree, RB-GP model reduces the unnecessary retransmission of information and channel conflicts.