Analytical Performance of Collaborative Spectrum Sensing Using Censored Energy Detection

The performance of collaborative spectrum sensing using censored energy detection is analyzed. Unlike the conventional energy detector that applies the measurements from the interested band directly to the decision-making process, the censored energy detector selects the measurements at different collaborating users by comparing them with two pre-determined limits before applying them to collaborative spectrum sensing. Both soft decision and hard decision rules are considered. Using the Neyman-Pearson criterion, analytical expressions for the probability of detection are derived and are verified by simulation. A simplified censored energy detector based on the Gamma approximation is also obtained. Using the derived results, it is shown that the censored energy detector outperforms the conventional energy detector when the optimum limits are used in the censoring. The performance gain depends on the decision rule, the operating signal-to-noise ratio and the number of collaborating users.