Subsection-average cyclostationary feature detection in cognitive radio

Spectrum sensing plays an important role in cognitive radios because the secondary users need to continuously monitor the spectrum for the presence of primary user. In this paper, we mainly investigated the cyclostationary feature spectrum detection in cognitive radios. Our analysis shows that cyclostationary feature detection requires partial information of the primary user and high computation cost although it is robust to interference in low SNR. We propose a novel strategy for spectrum sensing based on cyclostationary feature detection. Our new approach can effectively decrease the computational complexity and improve the performance of the inhibition of noise interference. At last, numerical results are provided in order to illustrate the advantages of our new technique.