Detection of the number of signals using a multiple hypothesis test

This work presents a novel approach to detect multiple signals embedded in noisy observations of a sensor array. We formulate the detection problem as a multiple hypothesis test. To control the global level of the multiple test, we apply the false discovery rate (FDR) criterion recently suggested by Benjamini and Hochberg instead of the classical familywise error rate (FWE) criterion. The proposed method is tested by the simulated data. Results show that the FDR-controlling procedure is more powerful than the FWE- controlling procedure. The performance improvement is most significant for a large number of signals and low SNRs.