Overlapping Channel Bonding Allocation for Dense
WLANs under Imbalanced Traffic Demands <Abstract> In recent IEEE 802.11 WLANs, an Access Point (AP) can use the Channel Bonding (CB) technique to increase the transmission bandwidth. Given limited bandwidth resources, it is not trivial to allocate channels for densely deployed APs to fairly meet their diverse traffic demands. In general, the non-overlapping channel allocation approach has been proven to be an ideal solution for maximizing the achievable throughput of WLANs using CB. However, we argue that the approach is no longer efficient because of two common situations: i) insufficient non- overlapping channels and ii) imbalanced traffic demands. Instead, the overlapping channel allocation is more suitable in such situations. In this study, we first apply a continuous-time Markov Chain (CTMC) model to analyze the effect of overlapping channel allocations for neighboring APs under unsaturated traffic conditions. Based on the analysis results, we propose a novel channel allocation algorithm based on the Bin-Packing Problem (BPP) that maximizes the fairness of the throughput-to-demand satisfaction ratio between APs. Through simulations, we demonstrated that the proposed algorithm can achieve near-optimal performance in various network settings. The algorithm also outperformed existing algorithms with the same overlapping channel approach. |