Traffic-aware channel assignment in wireless LANs

Abstract

Campus and enterprise wireless networks are increasingly characterized by ubiquitous coverage and high traffic demands. An efficient channel assignment of access points (APs) in these networks has the potential to greatly improve wireless LAN performance. Therefore, previous studies have investigated methods for determining effective channel assignments. Existing approaches, however, do not adapt to prevailing traffic conditions in the network. An analysis of wireless traces (e.g. [2]) shows that traffic volume in a wireless LAN can vary significantly across APs. Motivated by this observation, we investigate whether the quality of a channel assignment can be improved by incorporating observed traffic demands at APs and clients in wireless networks. Our results show that being traffic-aware can substantially improve the effectiveness of a channel assignment: in some cases it nearly doubles performance. We further apply traffic demand prediction algorithms to make traffic-aware assignment practical.