Partition function game for rate optimization in random interference networks

Interference is one of the main limiting factors in wireless communication networks for system performance nowadays. In this paper, we study a coalitional game approach for rate optimization in a random interference networks where multiple transmission pairs transmit concurrently. We model the rate optimization problem as a partition function game. All transmission pairs form separated coalitions, and the pairs cooperate to transmit within each coalition. An iterative coalition structure generation algorithm is proposed to obtain partitions. This algorithm allows each transmission pair to make individual decision whether to stay in or depart from its current coalition. We prove that the proposed iterative algorithm can achieve a Nash-stable partition. In each coalition, we adopt the duality features of broadcast channel (BC) and multi-access channel (MAC) to maximize the sum-rate. Simulation results show that the proposed algorithm can improve the network performance significantly compared with location-based static partition forms.