Novel PSD estimation algorithm based on compressed sensing and Blackman-Tukey approach

Spectrum sensing algorithms based on power spectral density (PSD) are an important category in spectrum sensing approaches. In this paper, a new PSD estimation method based on compressed sensing (CS) and the Blackman-Tukey (BT) approach is proposed. First, the low-rate measurement sequence of the received signal is obtained by CS. Second, the autocorrelation of the received signal is estimated from the low-rate measurement sequence. Then finally, the PSD of the received signal is computed using the BT approach. It is possible to estimate the wideband spectrum from a low sampling rate when using the proposed algorithm. Furthermore, the method does not reconstruct the received signal, which can significantly reduce complexity. Simulation results show that the NMSE performance of the proposed method is more excellent than the existing CS-MTM-SVD algorithm.