Game-Theory Based Interference Coordination In Hierarchical Heterogeneous Wireless Networks

The upgrade of wireless communication technology has facilitated the implementation of high-data-rate and ubiquitous access to network, and attracts more and more end users. The switching from voice traffic to data traffic has posed new challenges to the further improvement of system capacity and spectrum efficiency (SE), which accelerates the emergence of hierarchical heterogeneous networks (HetNets). However, the dense deployment of heterogeneous nodes results in severe co-tier and cross-tier interference due to the scarce frequency, making it important to effectively coordinate HetNets’interference. Recently, game-theoretic approaches have shown great increase in wireless technology areas, and have also dramatically improved system performance compared with conventional methods. Thus, interference coordination from the perspective of game theory shows significant values in both academic research and practical application. This thesis addresses the game-theoretic interference coordination in HetNets, covering both the non-cooperative case and the cooperative case.Firstly, heterogeneous D2D network is picked as an example, where the power control based interference management is studied using Stackelberg game. Macro cellular link and D2D links are respectively modeled as the leader and followers, with the objectives of maximizing their individual utilities. The existence and uniqueness of Stackelberg equilibrium are guaranteed through detailed theoretical analysis. During the update of leader’s price, the quality-of-service (QoS) requirements of both leader and followers are considered simultaneously by novelly introducing a weight factor. Simulation results verify the convergence of power and price iterations, and show that there exists a tradeoff between the performances of the leader and followers by dynamically changing the weight factor.Then, the cooperative interference coordination in small cell network is studied using coalitional game, where coalition members cooperate to suppress suffered interference. A novel coalition formation criteria called defection order is proposed to facilitate the cooperation, based on which a self-optimal coalition formation algorithm (SOCFA) is proposed and algorithm properties are further theoretically analyzed. Simulation results prove the convergence, low-complexity and fairness of proposed algorithm, and show the significant improvements compared with both the non-cooperative scheme and the conventional broadcasting based coalitional scheme.