Non-dominated sorting genetic algorithm for channel assignment in multiple radio interfaces with multiple channels

Mesh routers and mesh clients make up wireless mesh networks (WMNs), with the mesh routers connected to wireless devices serving as the network’s backbone. Wireless mesh routers can relay or transfer data in multiple hops, which lowers the cost of hardware and allows for easy deployment. These WMNs consist of multiple radio interfaces communicating on multiple channels (MRMCD). Hence, static channel assignment on MRMC WMSs plays a key role in the performance of WMSs. It aims at increasing packet delivery ratio, throughput, and network capacity while minimizing E2E delay. Researchers have developed numerous meta-heuristics searching-based channel assignments on MRMC WMNs; however, they struggle to balance multiple competing performance objectives, where improving one metric often degrades others, leading to uneven network performance and link starvation. In this article, we address these competing objectives for MRMCs using a non-dominated sorting-based algorithm with two variants (NSGA-II and NSGA-III). The developed MRMC is compared with traditional single objective-based optimization and demonstrates superior performance across most network metrics, with NSGA-III showing stronger results compared to NSGA-II.