Statistical Test Based on Finding the Optimum Lag in Cyclic Autocorrelation for Detecting Free Bands in Cognitive Radios

Cognitive radios sense the radio spectrum in order to find free bands for transmission. In this paper, a novel method is proposed for detecting free bands by exploiting a statistical test based on finding the optimum lag in cyclic autocorrelation of primary user signals within the framework of cyclostationarity theory. This approach has a satisfying detecting performance under low SNR conditions and is computational efficient. Meanwhile, due to signals´ cyclostationarity, the method also has a good anti-interference property which can be utilized to detect the primary user signals in the presence of interference. Finally, simulations are performed to verify those properties of the proposed method.