Reducing Location Information Overhead for Duplicate Address Detection in VANETs

• Main Authors: Kuan-Lan Tseng, 曾冠嵐
• Other Authors: Tsan-Pin Wang
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/7675pf

碩士 === 國立臺中教育大學 === 資訊工程學系 === 102 === In VANET (Vehicular Ad hoc Network), node moves at high speed and the network topology changes frequently. As a result, the connection between mobile nodes is sometimes unreliable. Nowadays, IEEE 802.11 (a)(b)(g)(n) have been developed, and are widely used in our lives. However, they may be unsuitable in high mobility network. To solve this problem, IEEE 802.11 adapts IEEE 802.11p, combined with the IEEE 1906 standards in order to expand the application of IEEE 802.11 to the VANET. Due to the node mobility in VANET, the connecting network changes frequently. Two or more ad-hoc networks may merge together during the movement on the road. The situation that nodes having the same IP address locate in the same network may occur consequently. In this case, the packets will be sent to the wrong destination. This thesis focuses on reducing location information overhead for duplicate address detection scheme in VANET. We propose a Grid-based Duplicate Address Detection (GDAD) scheme. In GDAD, the network is separated into numbers of grids according to the quadtree. To reduce the packet size in traditional Passive Duplicate Address Detection, we use the grid number to replace the GPS information in the packet. We adapt AODV routing protocol to transmit the packet in GDAD. AODV is a reactive routing protocol. It transmits packet with an on-demand method which can reduce the number of packets in the network. On the other hand, to confirm the situation when both of the IP address and grid number are conflict, we also propose a Grid Repartition Scheme. The performance evaluation shows that the packet size of GDAD is much smaller than the one in PDAD. Finally, we simulate our proposed GDAD in NS2 and the simulation results show that our Grid Repartition Scheme can improve the success detection ratio.