Intra-pulse Modulation Recognition Using Time-Frequency Features Based on Modified-B Distribution

In the electronic warfare environment, radars can be differentiated according to intra-pulse and inter-pulse     modulations. Detection of intra-pulse modulation with negative SNR is a topic of interest to researchers. In this paper separation of intra-pulse modulation with frequency and time-frequency methods is presented. Using this method, we can categorize different types of LFM, 4FSK, 2FSK, BPSK, and NM modulations. The algorithm of this method is based on characteristics and it is able to classify all radar signals from these types of modulations. To detect the   modulation, time-frequency characteristics based on the improved time-frequency transform, B, have been used. The innovation in this research, is the use of new characteristics of time-frequency distribution. The proposed algorithm uses time-frequency distribution to analyze radar signals. Dimension reduction is performed next, then for each     frequency the maximum time value is considered and the characteristics are extracted from signal. The presented   algorithm has 100% capability of separating radar signals for this number of intra-pulse signals up to -11dB of SNR whereas similar methods have less accuracy with SNR range between -5db to 5db.