Dynamic Pilot Allocation Scheme for Joint User Grouping and Alliance Game in Massive MIMO Systems

Abstract

Many researches demonstrate that cell-edge users are more susceptible to pilot contamination than the cell-center users in massive MIMO systems. Therefore, this paper proposes a dynamic pilot allocation scheme called Joint User Grouping and Alliance Game (JUG-AG) to mitigate pilot contamination. According to the user signal strength, the users are divided into two groups, namely A and B. Users with weak strength of received Base Stations (BSs) signals are recorded as group A, and the remaining users are group B. The users of group A use mutually orthogonal pilots, and the users of group B reuse the remaining orthogonal pilots by means of alliance game. In the alliance game for the users of group B, users are divided into several disjoint user sub-alliances, users belonging to different sub-alliances are allocated different orthogonal pilot sequences, and users in the same sub-alliance reuse the same pilot sequence. Compared with the existing pilot allocation schemes, the proposed JUG-AG scheme is more flexible and can be used for scenarios that all users are randomly distributed. Moreover, the algorithm can obtain the overall optimal solution through cyclic searching. The simulation results demonstrate that the JUG-AG scheme can effectively reduce the average Root Mean Square Error (RMSE) of user signal detection in the uplink and improve the average service rate of users.