Distributed Beamforming and Autonomous Participation Decision making in Cooperative CR Systems in Presence of Asynchronous Interference

Abstract

We propose a distributed beamforming for cooperative cognitive radio (CR) networks, that reduces the required feedback overhead in terms of sharing the channel state information (CSI) between the cooperating CR relays. Assuming the reciprocity of the channels of each cooperating CR node toward the CR receiver, as well as toward the primary receiver, the proposed beamforming technique only requires sharing the location information of the CR relays among the cooperating CR relays, rather than the global CSI. Reducing the required feedback overhead provides enhanced scalability of the cooperative CR network with lower deployment cost. In this paper, we also propose two autonomous decision making strategies that can help each CR user to independently decide whether to participate in the cooperative transmission or not. This participation decision is tackled from a game theoretic approach, by quantifying the reward and cost functions of each CR relay. The first proposed strategy, named regret testing-based strategy, is proved to asymptotically achieve an approximate Nash equilibrium state of the system. However, it has higher complexity and slower convergence time. So we also propose a suboptimal decision making strategy, named learning-based strategy, that has lower complexity and faster convergence time.