Self-organizing channel assignment for high density 802.11 WLANs

In a dense WLAN deployment, the interfering wireless access points (APs) need to efficiently share the spectrum, and hence, avoid low-performance experience of users due to high collision rates and long backoff overheads. In this paper, we aim to propose a channel assignment scheme in which APs can self-configure their channel choices to mitigate interference and thereby maximize network throughput. Aiming to minimize interference sum utility, a channel assignment problem is formulated as a non-convex optimization problem and solved using difference-of-convex-functions (DC) programming. Subsequently, two distributed algorithms are developed in which each AP independently adapts its channel selection to the optimal values over time by measuring the received interference in different channels. The convergence, complexity, and efficiency of the developed algorithms are studied by simulation results.