Cooperative Spectrum Sensing for Cognitive Radio Networks Based on Spectrum Estimates

Spectrum sensing is a major function within the cognitive radio system. In this paper, spectrum sensing using energy detection is considered and an accurate performance analysis is provided for both cases of single and cooperative sensing. The performance of the periodogram-based energy detector is studied for random signals employing a complex envelope. The contribution of this paper is twofold. New and accurate closed forms are developed for the performance measures which are the probability of false alarm and the probability of miss. The obtained mathematical models are supported by Monte-Carlo simulations. In addition, the obtained derivations are extended to include the performance of centralized cooperative sensing using decision fusion. The considered decision rules are Logical OR (LO), Logical AND (LA) and Majority Rule (MR). The obtained numerical results reveal that the probability of detection tends to increase when using LO and MR. On the other hand, LA yields the same receiver operator characteristic (ROC) curve irrespective of the number of cooperative users.