Compressed Sampling Signal Detection Method Based On Modulated Wideband Converter

In order to deal with the problem when the bandwidth of the passive radars and wideband digital receivers are getting bigger and bigger, causing that the traditional signal detection methods need a high sampling rate and a large number of sampling points. This paper extends the modulated wideband converter (MWC) which is recently proposed compressed sensing techniques to discrete-time domain in order to construct a wideband digital receiver. The MWC compressed sampling structure is applied to get the sub-Nyquist compressed sampling signal to detect whether the interesting signal exists or not. Then we can get eigenvalue decomposition of the covariance matrix of the compressed sampling data. According to linear distribution characteristics of Gaussian white noise covariance matrix eigenvalues, we can use the correlation coefficients of these eigenvalues to construct signal detection criterion. The simulation results show that the proposed signal detection method in this paper obtains a higher signal detection success rate in low signal to noise ratio (SNR) with fewer sampling data compared with conventional signal detection method.