Traffic-aware channel assignment in enterprise wireless LANs

Abstract

Campus and enterrprise wireless networks are increasingly characterized by ubiqitous coverage and rising traffic demands. Efficiently assigning channels tp access points (APs) in these networks can significantly affest the performance and capacity of the wireless LANs. Several research studies have tackled this issue. However, state-of-the-art approached assign channels statically without considering prevailing traffic demands. In this thesis, we explore whether the quality of a channel assignment can be improved by incorporating observed traffic demands at APs and clients into the assignment process. Using extensive simulations over publicly available wireless traffic traces, as well as synthetic settings, we show that being traffic aware can substanially improve the overall quality of a channel assignment. We also identify several situations where traffic awareness is likely to be of great help, and the several where the benefit is minimal. We also address several practical issues in deploying our scheme in real WLANs, including measuring an interference graph, collecting traffic demands, and predicting future demands based on historical information. On the whole, our approach is similar yet very effective. It can be incorporated into existing WLANs with little to no modification to the wireless nodes or infrastructure. We present an implementation of our assignment scheme for modest scale wireless testbeds. Our experiments with this implementation show that traffic-aware assignments can significantly improve the network performance under a wide range of real network configurations.