Credibility test for blind processing results of BPSK using approximate entropy

A credibility test method based on approximate entropy (ApEn) is proposed to evaluate the blind processing results of BPSK signals in Cognitive Radio (CR). In the absence of a priori knowledge for the signals, the reference signals were constructed depending on the modulation recognition results by the specific modulation recognition scheme at first. Then the modulus of the cumulative sum (CUSUM) between the observed signals and the reference ones were calculated. By analysis of the differences between ApEns (DApEn) of the CUSUM and its ascending ordered form, the credibility test for BPSK signals is boiled down to detect whether the DApEn is 0 or greater than 0. The threshold for detection is set based on Bootstrap method. Simulation results show that the proposed method can be used to verify the credibility for blind processing results of BPSK signals under low signal-to-noise ratio (SNR).