Distributed schemes for fair throughput in infrastructure-based IEEE 802.11 mesh networks

This paper studies throughput fairness among different basic service sets (BSSs) in infrastructure-based IEEE 802.11 mesh networks, where inter-BSS interference is unavoidable because of the difficulty in frequency and coverage planning and the limited number of non-overlapping frequency channels available for IEEE 802.11 networks. Two approaches are proposed for achieving fair throughput among the co-channel BSSs: (i) dynamic network allocation vectors (DNAV), which dynamically adjusts the co-channel interference by temporarily stopping transmissions in some BSSs using network allocation vectors (NAVs), and (ii) forced handoffs (FHO), which forces mobile stations (MSs) to handoff between neighboring BSSs. An optimization problem is formulated as well as a distributed scheme is proposed for each of the approaches for achieving the required throughput fairness while keeping high transmission throughput in the BSSs. Our results show that (i) the DNAV scheme can always achieve the required throughput fairness; (ii) performance of throughput fairness using the FHO scheme depends on coverage overlapping between the BSSs, and when there is strong coverage overlapping among the BSSs, the scheme achieves a higher throughput than the DNAV while achieving the required throughput fairness; and (iii) the converged throughput performance of the proposed distributed schemes is very close to the corresponding optimized one.