The design and evaluation of an opportunistic multicast routing protocol

Abstract

This thesis aims to examine the efficiency of opportunistic multicast routing in Wireless Mesh Network (WMN). Specifically, an opportunistic multicast routing protocol has been developed in this work in order to gain better performance in WMNs compared to traditional multicast routing protocols. The number of transmissions in the network to deliver a packet from a source to group of destinations has been taken as measure for the performance of the protocol. Less number of transmissions in the network to deliver a packet to group of destinations is an indicator for a better performance. The work has been divided into two parts. In the first part, a proposal of the first version of opportunistic multicast routing protocol (OMRPv1) will be explained. By calculating the expected number of transmissions in the network, the efficiency of the protocol will be examined. OMRPv1 will be compared against a traditional protocol, non-opportunistic protocol, to confirm its better performance in the network. Then, a second version of our protocol will be introduced, OMRPv2, as a solution for some problems encountered in the first version. The comparison between OMRPv1 and OMRPv2 will show the better performance achieved with the newer version. In the second part, QualNet simulator will be used to simulate the protocol OMRPv2. In this part, not only the number of transmissions in the network will be considered, but also other metrics like end to end delay and delivery rate will be taken into account. The overhead of the protocol will be examined as well. Observation of the results from the simulation will lead to the third version of the protocol (OMRPv3). OMRPv3 will be simulated as well to show the improvement in the performance over the previous version OMRPv2. Finally, OMRPv3 will be compared to the best opportunistic multicast routing protocol in the literature, Beacon-less Geographic Routing for Multicast Applications (BRUMA). For that reason, BRUMA will be simulated in QualNet as well. For fair comparison, few modifications will be added to BRUMA. Then, the two protocols will be compared to each other using many metrics. Delivery rate, end to end delay, and number of transmissions in the network to serve the multicast groups are some of the metrics that will be used to compare the two protocols. Results show that our protocol, OMRPv3, outperforms BRUMA in many of these metrics.