Convex Sensing-Reporting Optimization for Cooperative Spectrum Sensing

We consider the cooperative spectrum sensing problem in cognitive radio with energy detection. Secondary users with non-identical sensing channels make 1-bit sensing decisions that are reported to the secondary base station over non-identical noisy fading channels. The base station has reporting channel knowledge and combines the decisions with an M-out-of-K rule. We allow the secondary users to trade sensing time samples for additional reporting time slots to increase the reporting signal-to-noise ratios. We derive the false alarm and missed detection probabilities as functions of the secondary sensor decision thresholds and the durations for sensing and reporting. Furthermore, we bound these probabilities and impose practical constraints that enable convex optimization to minimize the false alarm probability for a target missed detection probability. We compare knowing the instantaneous reporting channels for optimization with knowing the average channels. Allowing secondary users to trade sensing time samples for additional reporting time slots is shown to significantly improve sensing performance, even with poor sensing and reporting channels and a small number of secondary users.