Research Of Resource Allocation Algorithms In 802.11 Wireless Networks

In recent years, the IEEE 802.11-based wireless local area networks (WLANs) have been growing at exponential rate. Under the widespread deployment, such as in offices, airports, pub-lic buildings etc., the density of Access Points(APs) and user stations(STAs) has increased sub-stantially. In such dense wireless networking environment, the management entity specified by default 802.11 protocol, including user association, AP channel allocation and power control schemes, are shown to be ineffective in combating the co-channel interference problem between neighboring nodes due to high node density. This issue is further exacerbated by the imbalance load problem across different APs due to the non-uniform distribution of STAs, leading to serve performance degradation to end users.This thesis attempts to address these issues by proposing game theoretical solutions for the user association, channel allocation and power control problems.Firstly, the user association problem is considered whereby the objective is to balance the traffic load across different APs. We analyze the metric for measuring the traffic load level of APs in dense 802.11 WLANs, with which a game theoretic model is developed for the user association problem. Both centralized and distributed algorithms are proposed for the user association game. We provide simulation results show that the association algorithms can balance the traffic load across APs effectively and outperform the existing scheme in terms of fairness and load balance.Secondly, we extend the model to the case whereby the number of orthogonal channel is lim-ited. As a result, the user association problem has to be considered with AP channel allocation, which is formulated as a non-cooperative game. We prove the existence of the Nash equilibrium, and derive price of anarchy and the fairness index for the game. The analytical results are vali-dated with simulation study, which show that the proposed algorithm can effectively assign the channels to APs while control the association of users to balance the load of APs on different channel.Thirdly, we consider the power control problem in dense 802.11 WLANs for further im- provement of the performance of the networks. A two-step power control algorithm is proposed to alleviate the co-channel interference between APs, which can preserve the carrier sensing symmetry between different nodes and thus avoids the starvation problem. Simulation results are provided to show that the algorithm can reduce the congestion level of users effectively.Finally, we discuss the potential applications of the algorithms proposed in this thesis and point out the possible directions for future research.