Efficient Wideband Spectrum Sensing Based on Compressive Sensing and Multiband Signal Covariance

This paper a novel and efficient wideband sensing algorithm based on compression and reconstruction of the covariance matrix of the signal from the covariance matrix of the acquired samples is proposed, it allows users to sense the cognitive spectrum without a priori knowledge of signal characteristics in the radio environment. Simulation results show that the proposed method allows estimating the spectral covariance matrix of signal and from it can perform efficiently the spectrum sensing, improving performance sensing according to the detection probability, false alarm and probability of failure detection, compared with spectrum sensing algorithms based on wideband energy detection, which works at rates above or equal to the Nyquist sampling rate.