p-Value based cooperative multiband spectrum sensing for cognitive radio

Abstract

In this paper, we study an implementation of the Benjamini-Hochberg's procedure for cooperative multiband spectrum sensing in cognitive radio. We propose to use equal gain combining on forwarded p-values from cooperative cognitive radio (CR) users as the test statistic at the fusion center. By doing so, the distribution of the test statistic under the null hypothesis at the fusion center does not vary with different number of samples and types of local detectors. Simulation results show that the familywise miss detection rate (FWM) significantly decreases as the number of cooperative CR users increases. This means that the interference to a primary user due to the decision errors in spectrum sensing significantly decreases. Thus, p-value based cooperative multiband spectrum sensing using the Benjamini-Hochberg's procedure is well motivated.