Skip to Main Content
In cognitive radio (CR) cooperative networks, the applicable relay scheduling and channel access directly affect the system. However, in practical system, the CR sensor is bound to sensing error. This may lead to low system performance and interfering with primary users. In this paper, we present a learning based centralized scheme for joint relay scheduling and channel access scheme in CR cooperative networks. Specifically, we formulate the joint relay scheduling and channel access problem as a partially observable Markov decision process (POMDP) system, where the most likely channel state is derived by a learning process. The optimal policy is derived by solving the problem. Extensive simulation results show the effectiveness of our proposed scheme.