An enhanced multicast-based optimized link state routing for wireless community network

Abstract

Wireless community networks (WCNs) are a solution for people who are living in some areas facing difficulties for accessing the Internet because no Internet service providers (ISPs) are providing them with the service due to the long distance, the high cost of infrastructure, and the less number of people in these areas. The current routing protocols for WCNs have two performance issues. The first issue is routing stability. Because of the heterogeneous characteristics of links, link quality should be considered as one of the main metrics used to control the routing of packets. However, current routing techniques depend on the shortest path as the main metric to control the routing of packets which results in non-stable routes. Non-stable routes affect communication speed which is a main requirement for large scale of real time applications. The second issue is routing scalability. The scalability is more challenging in the presence of both large number of nodes and mobility. As current routing protocols are inefficient when faced with the dynamic changes and poor links that occur in real-life and self-managed deployments. This results in too much overhead during communications due to flooding as most of the current routing protocols uses unicast traffic. In this research, the ad hoc routing protocol, optimized link state routing (OLSR) is selected and enhanced so that it can meet the standards of efficiency in terms of stability and scalability. OLSR is enhanced through three phases. The first phase is the multicasting expansion where Multicast traffic is expanded to the OLSR routing protocol in WCNs in order to decrease the overhead caused by flooding as OLSR uses unicast traffic. The second phase is the multipoint relay (MPR) selection based on analytical hierarchical process (AHP). Multiple criteria are taken into account simultaneously in the Multi-Criteria Decision Making (MCDM) method to create a flexible decision making process. Multiple metrics can be weighted according to MCDM: AHP. Each node establishes an MPR set based on a single cost determined with the given metrics. The third phase is a composite metric for optimal route selection. The composite metric is proposed using multiple parameters in order to ensure good knowledge of the status of links that can guarantee picking the most stable links in the network. The aim of the new proposed metric is to make finding the best routes extremely easier with the dynamic topology of WCNs. In addition, it aims to avoid the use of hop count metric which is used in the OLSR protocol and is not suitable to the dynamic link characteristics of WCNs. The new proposed routing protocol is developed using C++ programming language under the NS-2 simulator. The performance of the proposed routing protocol is measured using four performance metrics: average end-to-end delay, network control overhead (NCO), packet delivery ratio (PDR), and energy consumption in terms of network density and traffic load with varying mobility speeds. The proposed routing protocol outperforms the OLSR protocol in terms of average end-to-end delay, NCO, and PDR by 5%, 11%, and 12% respectively. While, the energy consumption for the proposed routing protocol is approximately similar to the standard OLSR protocol.