Research On Recognition And Parameter Estimation Of OFDM Radar Signal

With the continuous application of the new radar system in modern electronic warfare,the research of the new radar system has become a hot research topic.Orthogonal Frequency Division Multiplexing(OFDM)radar is a new system radar with high spectrum utilization,good anti-interference ability and low interception performance,which brings great challenges to the electronic reconnaissance of OFDM radar.In accurate electronic warfare,it has become an indispensable part to effectively identify the modulation type of OFDM radar signal and accurately estimate the parameters of radar signal.In view of above problems,the OFDM radar signal is used as the object to study the modulation recognition and parameter estimation of OFDM radar signal.The specific research work is as follows:1.Three modulation types of OFDM radar signal model are studied.The time domain,frequency domain and time-frequency domain of each OFDM radar signal are analyzed.Four intra-pulse features are extracted from OFDM radar signal.These lay a foundation for subsequent OFDM radar signal recognition and parameter estimation research.2.The recognition method of OFDM radar signal based on TPOT and LIME is proposed.This method can not only accurately identify the modulation type of OFDM radar signal,but also make a correct interpretative evaluation of whether the predicted signal is recognized correctly.In order to solve the problem that the artificial design features need to be extracted in the recognition process and the recognition effect of OFDM radar signal is poor under the complicated noise environment,A convolutional neural network structure suitable for OFDM radar signal recognition is proposed.The network structure uses the one-dimensional convolutional neural network to learn the time domain and frequency domain features of the OFDM radar signal and fuse the learned features to realize recognition of modulation types of the OFDM radar signal.The effectiveness of the methods are verified by simulation.3.Based on the accurate recognition of the modulation types of OFDM radar signal,the corresponding modulation parameter estimation algorithms are proposed.For the LFM-OFDM radar signal,the fractional Fourier transform method is used to estimate the modulation rate and the number of subcarriers.After estimating the modulation rate of the signal,the signal is de-modulated and then the carrier frequency of the signal is estimated by analyzing the frequency spectrum of the de-modulated signal.For the MLFM-OFDM radar signal,LSD algorithm is used to analyze the time-frequency diagram of the signal,thus the modulation rate and the number of segments of the signal are estimated.After estimating the modulation rate of the signal,the signal is de-linearized and then the carrier frequency and the number of sub-carriers are estimated by spectrum analysis of the de-modulated signal.The modulus square analysis of LFM-OFDM and MLFM-OFDM radar signal is used to estimate the subcarrier frequency interval of signal.For the PSK-OFDM radar signal,the wavelet transform analysis method is used to perform two discrete wavelet transforms on the signal to realize the code rate estimation of the signal.The effectiveness of the above algorithms are verified by simulation.