Observations on Using Simulated Annealing for Dynamic Channel Allocation in 802.11 WLANs

Reliable channel selection for each Access Point (AP) is essential in setting up and operating densely deployed 802.11 WLANs. The aim of the channel selection is to provide efficient reuse of spectrum and therefore minimize interference and improve user experience. This paper highlights the potential of using Simulated Annealing (SA) to solve the channel selection problem in single-hop multi-AP 802.11b/g networks. The objective is to develop a framework to gauge how well the SA performs in a wireless communication environment. This involves the design of neighboring state transition scheme, annealing schedule selection and the analysis of their impact on the system performance. The simulation results show that the proposed algorithm solves the channel allocation problem and finds assignments very close to the known optimal solutions for many different network topologies.