Spectral covariance for spectrum sensing, with application to IEEE 808.22

Despite the shortage of available frequency spectrum, recent studies have shown that the actual usage of the allocated spectrum is scarce. The IEEE is developing the 802.22 standard for spectral reuse in TV bands that uses cognitive radio (CR) technology. One of the essential and challenging features of CR is spectrum sensing. This paper proposes a novel spectrum sensing algorithm using spectral covariance of the received signal. The proposed spectral covariance sensing (SCS) algorithm exploits different statistical correlations of the signal and noise in the frequency domain. Test statistics are computed from the covariance matrix of spectrogram and compared with the decision threshold. Detection performance is theoretically analyzed and verified through extensive simulation according to the IEEE 802.22 requirements. We show that SCS achieved 3dB better sensitivity to pilot location detector with the same sensing time. It is also shown that SCS is very robust to the noise uncertainty, which is one of the critical performance measures of spectrum sensing. The results of the paper suggest that SCS is an efficient and viable sensing scheme for IEEE 802.22 systems.