A Novel Geographic Routing Strategy over VANET

• Main Authors: Yan-Bo Wang, 王彥博
• Other Authors: 李維聰
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/71117492461810371436

碩士 === 淡江大學 === 電機工程學系碩士班 === 98 === In recent years, thanks to the development and popularization of wireless network technologies, the issue of Vehicular Ad-Hoc Network (VANET) has received great attention, and more and more VANET-related researches have been brought up. Generally speaking, the biggest difference between VANET and traditional Ad Hoc Network is the velocity of carriers because in VANET, the velocity of vehicles the carriers is much higher than those in traditional Ad Hoc. Therefore, it would be a great challenge to forward data efficiently in VANETs and many researches proposed have focused on the development of routing protocols. The current proposed routing protocols are all assumed to simulate in a distributed and ideal environment. As for the complex geographic environments, like urban scenarios, extra amendments must be needed to improve the efficiency of the routing protocols. In traditional Mobile Ad Hoc Network (MANET), there are corresponding routing protocols, such as DSDV, AODV, DSR, to assist the mobile node to find the path from source to destination, when the mobile node want to send data to destination. Accompany with the velocity of mobile nodes become higher, such like the vehicles in VANET, the traditional Ad Hoc routing protocols are inadequate. Therefore, the geographic position-based routing protocols are proposed to support the feature of the mobility of nodes. When the geographic position-based routing protocols are proposed, it focuses on solving the problem only that is the feature of the mobility of nodes. It use the Global Positioning System (GPS) to obtain the geographic information, and the information support the routing protocol to make routing decision when the routing protocol only has the local topology information. But it is also inadequately in some scenario, such as urban scenario. Thus, the main purpose of this paper is to design a suitable routing protocol for urban scenarios with better performance and adaptability. For this reason, Greedy on Straight Roads and Predictive at the Intersections (GSPI) routing protocol is proposed: to use greedy mode on straight roads, and to use predictive mode at the intersections. In greedy mode, we choose the next hop according to the weight value (Weight, Wi) that combines the distances and multi-rate. In predictive mode, we predict the direction of the vehicles to determine the next hop. The simulation results of Network Simulator (NS-2) reveal that our proposed algorithm indeed proves its feasibility.