Topology preservation and control approach for interference aware non-overlapping channel assignment in wireless mesh networks

The Wireless Mesh Networks (WMN) has attracted significant interests due to their fast and inexpensive deployment and the ability to provide flexible and ubiquitous internet access. A key challenge to deploy the WMN is the interference problem between the links. The interference results in three problems of limited throughput, capacity and fairness of the WMN. The topology preservation strategy is used in this research to improve the throughput and address the problems of link failure and partitioning of the WMN. However, the existing channel assignment algorithms, based on the topology preservation strategy, result in high interference. Thus, there is a need to improve the network throughput by using the topology preservation strategy while the network connectivity is maintained. The problems of fairness and network capacity in the dense networks are due to limited available resources in WMN. Hence, efficient exploitation of the available resources increases the concurrent transmission between the links and improves the network performance. Firstly, the thesis proposes a Topology Preservation for Low Interference Channel Assignment (TLCA) algorithm to mitigate the impact of interference based on the topology preservation strategy. Secondly, it proposes the Max-flow based on Topology Control Channel Assignment (MTCA) algorithm to improve the network capacity by removing useless links from the original topology. Thirdly, the proposed Fairness Distribution of the Non-Overlapping Channels (FNOC) algorithm improves the fairness of the WMN through an equitable distribution of the non-overlapping channels between the wireless links. The F-NOC is based on the Differential Evolution optimization algorithm. The numerical and simulation results indicate that the proposed algorithms perform better compared to Connected Low Interference Channel Assignment algorithm (CLICA) in terms of network capacity (19%), fractional network interference (80%) and network throughput (28.6%). In conclusion, the proposed algorithms achieved higher throughput, better network capacity and lower interference compared to previous algorithms