Wideband Spectrum Sensing Based on Nested Sampling for Noncircular Signals

Spectrum detection using Nyquist sampling is a very effective wideband spectrum sensing scheme. However, the current Nyquist method needs high rate of sampling to implement the ideal detection probability when the amount of sub-bands gets more. At the same time, for the noncircular signal, the current wideband spectrum sensing technology does not consider that the conjugate autocorrelation information of the noncircular signal is not zero, which results in the low detection performance of the signal. Consequently, this paper presents a scheme of wideband spectrum detection using nested sampling of noncircular signals, which has the advantages of low sampling rate, high compression ratio and high detection probability. This method estimates the autocorrelation function and conjugate autocorrelation function of wideband signal by using nested sampling, obtains two kinds of power spectrum of signal by FFT, and then detects the sum of energy of each sub-band. The MATLAB results show that the detection probability of the new scheme is higher than the traditional scheme without considering the conjugate autocorrelation information of noncircular signals.