A Novel Spectrum-Sensing Method Based on Maximum Cyclic Autocorrelation Selection for Cognitive Radio System

In the dynamic spectrum access (DSA) type of cognitive radios, secondary users need to detect the signals from the primary system prior to communicating in the sharing band. Hence, spectrum sensing is an important function for DSA. Key requirements for spectrum sensing in realistic radio environments are stable performance in detecting the primary signals as well as robustness against noise uncertainty at the secondary device or interference signals from other secondary systems. This paper proposes a novel spectrum-sensing method based on maximum cyclic autocorrelation selection (MCAS), which exhibits good detection performance and robustness against noise uncertainty and interference with low computational complexity. Our MCAS-based spectrum-sensing method is used to detect whether the primary signal is present or not, by comparing the peak and non-peak values of the cyclic autocorrelation function (CAF). Our MCAS-based spectrum-sensing method does not require noise variance estimation. Furthermore, it is robust against noise uncertainty and interference signals. Through computer simulations, we found that our method performs as well as or better than conventional sensing methods and is robust against noise uncertainty and interference signals. Therefore, it could be a practical candidate in realistic radio environments.