A Cooperation Strategy Based on Bargaining Game for Fair User-Centric Clustering in Cloud-RAN

Coordinated multi-point (CoMP) has emerged as a promising concept for improving the overall network throughput, where each user is attached with a group of radio heads called a cluster. In this letter, a user-centric clustering scheme in cloud radio access network is proposed to consider the fairness between users while choosing the serving clusters. The CoMP clustering strategy is formulated as a cooperative bargaining game problem and resolved using Nash bargaining solution (NBS). First, the users are grouped into coalitions using a Hungarian method. Then, a two-band partition algorithm is developed for the two users in each coalition to bargain their attached radio heads so that the NBS utility is maximized. Simulation results show that the proposed algorithm can achieve a considerable rate gain and fairness improvements compared to the existing schemes.