Collaborative Change Detection for Efficient Spectrum Sensing in Cognitive Radio Networks

This paper proposes a collaborative change detection scheme for cognitive radio networks, where secondary users can collaboratively and efficiently detect the departure and arrival of the primary user´s signal. Specifically, we consider a system where each of the secondary users in the network can make a local detection in each time frame and forward its decision to the fusion center where a CUSUM-based change detector is used to efficiently and reliably detect the presence of the primary user. The change detector is developed to maximize the throughput of the secondary network subject to constraints on the interference imposed on the primary user. In accordance with the considered system model and the adopted detection scheme, this paper also analyzes the average throughput of the secondary network and the average interference time to primary users. The simulation results show that the proposed spectrum sensing scheme outperforms the fixed-sample-size detector in terms of the average throughput of the secondary network.