A Study Of AP-Based Geography Routing Algorithm For VANET In Urban Environment

As the development of the vehicular ad hoc network, researches on routing protocols of VANET in urban environment have received more and more attention. One of the most important features of VANET is the high mobility of vehicular nodes, which leads to a complicated and rapidly changing network topology and unstable communication channels. Both of those take great difficulties of designing high-performance routing protocols. In the real world application, inaccurate information acquisition and excessive routing cost have been commonplace in existing routing protocols. To solve this problem, an adaptive routing protocol in VANET is proposed in this paper, which is called the AP-based geography and landmark overlays routing. By the help of the navigation device, the protocol uses two mechanisms to insure a reliable and stable routing. The one is the AP-based estimate strategy of the road density, which takes advantage of the proper unit in VANET, the AP, to gather the distribution information of vehicles in order to avoid the low accuracy and the high cost of existing density estimate strategies; the other is the on-demand mechanism of the neighbor discovery, which gives up periodic broadcast messages and uses a passive information acquisition instead, so as to decrease the redundancy and enhance the real time performance.In this paper, the network simulator NS2 is used to evaluate the research findings, while the mobility modeling simulator VanetMobiSim is used to generate the real motion trail of vehicles in VANET. By comparing the AGLOR protocol and the existing LOUVRE protocol, the AP-based estimate strategy turns out to have higher accuracy and stability, while the number of routing messages in AGLOR is proven to have higher scalability and be less than which in LOUVRE. Also, the AGLOR protocol shows a lower loss rate. Consequently, in VANET, the proposed protocol in this paper can provide a better performance.