Soft Combining for Cooperative Spectrum Sensing over Fast-Fading Channels

We propose soft combining methods for cooperative spectrum sensing in cognitive radio networks over fast-fading channels. In fast-fading environment, it is difficult to obtain the exact probability distribution of the total received energy during an observation period. Therefore, we apply the mixture of gamma (MoG) approximation to obtain the distribution function of the total received energy. By exploiting the approximated distribution, we derive a soft combining method that reduces the probability of missed detection for a given probability of false alarm based on the Neyman-Pearson criterion. In addition, we propose a simplified soft combining method by using the dominant component of the MoG distribution. Simulation results show that the proposed combining methods outperform the conventional methods based on the central limit theorem, the equal gain combining, and the SNR weighted combining.