Document Type : Original Article
Authors
1
Assistant Professor, Department of Computer, Shahr-e-Qods Branch, Islamic Azad University, Tehran, Iran
2
Master's degree, Computer Department, Faculty of Technology and Engineering, Shahr-Quds Branch, Islamic Azad University, Tehran, Iran
Abstract
Routing algorithms in VANETs are classified into two categories: topology-based algorithms and position-based algorithms. Position-based algorithms are more suitable for VANETs due to their scalability. In these algorithms, the nodes obtain information about their neighbors using Hello messages, and then based on the information obtained from the neighbors, the next hop towards the destination is selected. The Geo-LU (Geographical Link Utility) algorithm is a position-based routing algorithm in which nodes collect the information of two-hop neighbors in addition to one-hop neighbors information. In this way, they expand their local view in selecting the next hop towards the destination. According to the information obtained from one-hop and two-hop neighbors, a pair consisting of one one-hop neighbor and one two-hop neighbor is selected towards the destination. In this algorithm, among the candidate pairs, a pair is selected which, while having a shorter distance to the destination, has a better link quality (a link with a higher delivery rate) and more bandwidth. In the Geo-LU algorithm, one of the most important challenges of VANETs, which is the short lifetime of links due to the mobility of nodes, is not considered in the selection of the next pair. Therefore, in this paper, the efficiency of the Geo-LU algorithm is improved by considering the lifetime of links when selecting the next pair. In this way, the proposed method can perform better than the Geo-LU algorithm against the mobility of nodes and the loss of connections in VANETs. The proposed method has been simulated using NS2 and several experiments have been conducted to check and evaluate its performance. The simulation results show that the proposed method compared to the Geo-LU protocol in scenarios with different density of vehicles increases 3% the packet delivery ratio and decreases 6% the end to end delay. Also the proposed method in scenarios with different speed of vehicles increases 5% packet delivery ratio and reduces 20% end-to-end delay.
Keywords
Main Subjects
)