A Study On The Digital Modulation Identification Technologies Based On Wavelet Transform

The digital modulation identification technologies are rapidly developing and applied to many fields including electronic war, software radio, wireless communication supervision, etc.Three main classification technologies of modulation type are presented, and their advantages and disadvantages are analyzed in this paper at first. Then the wavelet transformation theory is introduced. Because wavelet transform has good localizing time and frequency characteristics, and multi-resolution performance, it is an effective technique for extracting the transient characteristics from different digital modulation signals without any prior knowledge. The wavelet transformations of four kinds of different digital communication signals are deduced theoretically, and the modulation identifier is constructed based on these results. The inter-class and intra-class modulation type classification is investigated under two kinds of signal-to-noise ratio environments. The classifier carries out the inter-class modulation identification by comparing the variances of wavelet transform amplitude with the thresholds, and the intra-class identification by the values of wavelet transform amplitude and peak.Simulation results from seven kinds of common digital modulated signals show that the proposed classifier has high accuracy and low computational complexity. It is able to distinguish the different modulation types with average success rate 92.6% over the SNR of 5dB in the low SNR classifying method and 94.2% over the SNR of 15dB in the multi-SNR classifying method respectively.