A distributed interference control scheme in large cellular networks using mean-field game theory

Abstract

Considering a dense cellular network with a large number of base stations, this paper proposes an intercell interference control scheme using mean field game theory. Mean field game theory has been proven to be a more tractable technique than the traditional game theory. In the process of formulating the mean field game, we exploit statistical physics results to decouple the complex system of micorcells into several entities. These entities (players) are symmetric in terms of their action sets, and are interdependent by a consistent condition such that the interaction among them can be controlled. Moreover, Each entity is made capable of collecting brief and sufficient information about the system. Game theory and economic concepts are then used to decide the best transmit power. Fairness and bit error rate have been captured in the model. In different network topologies, the simulation results show that the proposed scheme achieves much better tradeoff between spectral efficiency and energy efficiency compared to different frequency reuse patterns.