A Traffic-Demand-Aware Collision-free Channel Allocation for Multi-channel Wireless Mesh Networks

Using multiple channels in wireless mesh networks (WMNs) can reduce collision and interference, which improves network performance. Reasonable channel allocation is an effective method for eliminating collisions in WMNs. In this paper, we aim to achieve collision-free channel allocation that is suitable for the traffic patterns in the target networks. Given a traffic-demand matrix of a network, we propose the Traffic-Aware Centralized Channel Allocation (TACCA) method to satisfy traffic demand without collisions in multi-channel WMNs. By incorporating a Carrier Sense Multiple Access aware (CSMA-aware) interference model, we formulate an optimization problem as Mixed Integer Linear Programming (MILP), which generates an optimal channel allocation when solved by an efficient solver. In TACCA, routing paths are consistently selected with satisfying capacity. Simulation studies show that TACCA can drastically improve performance for multi-channel WMNs by adjusting channel assignment incorporating the given traffic patterns.