Load Balanced Ad-Hoc On-Demand Routing Based on Weighted Mean Queue Length Metric

Uneven load distribution is one of the main issues in wireless mesh networks (WMNs) which leads to an inefficient use of the network resources, increases latency in some routes, and decreases the average packet delivery ratio. To tackle this problem, we propose two load balancing algorithms based on the Ad-hoc On-demand Distance Vector (AODV) routing protocol for WMNs, which called AODV-Mean Queue (AODV-MQ) and AODV-Weighted Mean Queue (AODV-WMQ). The load balancing algorithms provide an essential solution for dealing with the load congestion in mesh and gateway nodes. The main idea is to consider the average queue length of each node for forwarding or dropping the route request packets. In comparison to the AODV and AODV-Current Queue (AODV-CQ), simulation results show that the proposed load balancing algorithms yield significant performance gains in terms of the packet delivery ratio, End-to-End delay, and traffic load balancing on nodes.