Compressive sensing signal detection method base on numerical characteristics of sampling value

The samples obtained through compressive sensing effectively maintain structures and information of the original signal, so detection tasks of the original signal could be solved by directly processing the sampling values without reconstructing the original signal. Currently, the partial signal reconstruction method is mostly used for signal detection, which usually gets bad detection results under low SNR. For this case, this paper proposes a detection method based on the numerical characteristics of sampling values. According to the different characteristics of the expectation of sampling values under different hypothesis, detection is accomplished by using the deviation of the actual sampling values from the expectations under corresponding hypothesis as criterion. Experiments show that compared with conventional method the one proposed in this paper obtains a higher success rate of detection with fewer samples under low SNR.