Investigation on Signal Modulation Recognition in the Low SNR

Because of a low Signal-to-noise ratio (SNR), generally the recognition rate and recognition efficiency of signal modulation type are low, but the algorithm is complex. The recognition method of wide-band signal modulation type in the low SNR is studied. Based on the analyzing the characteristic of signal in time domain, frequency domain and power spectrum, three characteristic with fine classification feature are selected, and single k-NN Nearest-neighbor pattern classifier is adopted, and high recognition-rate of several wide-band signal modulation types is achieved in low SNR. The computer simulation results show that this method possesses perfect ability of modulation recognition to signals in a SNR of above 3dB, with an average recognition-rate higher than 99%. In addition, the design of recognition system is simple. It will have significant application value in detecting wireless signals.