QoE-Aware Downlink User-Cell Association in Small Cell Networks: A Transfer-matching Game Theoretic Solution With Peer Effects

The user-cell association mechanism is one of the important research topics for radio resource management in heterogeneous wireless networks. Existing studies mainly concerned the physical performance such as throughput and SINR, and ignore the upper layer users demand. For avoiding blindness of pursuing higher data rate and achieving more rational user-cell association, a novel method is proposed to associate users with heterogeneous traffic to small cell base stations (SBSs) based on user quality of experience (QoE). The user-cell association problem is formulated as a distributed transfer-matching game between SBSs and users to address the sum-QoE maximization problem. Furthermore, an effective distributed cooperative transfer-matching self-optimizing algorithm, roulette transfer matching algorithm, is designed for exploring the stable point of the game. It is proved that the proposed algorithm can converge to a stable solution and find the optimal two-sided transfer matching. Numerical experiments are presented to validate the proposed scheme and show the improvement for the system performance and fairness.