Locally best invariant test for multiple primary user spectrum sensing

We consider multi-antenna cooperative spectrum sensing in cognitive radio networks, when there may be multiple primary users. A noise-uncertainty-free detector that is optimal in the low signal to noise ratio regime is analyzed in such a scenario. Specifically, we derive the exact moments of the test statistics involved, which lead to simple and accurate analytical formulae for the false alarm probability and the decision threshold. Simulations are provided to examine the accuracy of the derived results, and to compare with other detectors in realistic sensing scenarios. From the considered simulation settings, performance gain over several known detection algorithms is observed in scenarios with relatively low signal to noise ratio.