Frequency Hopping Data Link Signal Modulation Identification And Parameter Estimation Technology

Data link is an important part of information warfare,and frequency-hopping system data link has become a research hotspot and has been widely used in many countries because of its strong anti-jamming,anti-spoofing and anti-interception performance.At present,with the improvement of scientific research in the field of communication,the diversification of wireless communication signal system and the increasingly large signal source,resulting in the increasingly bad wireless communication environment,which has brought huge challenges for communication countermeasures.Therefore,how to effectively identify and estimate the modulation mode and parameters of data link signal under the condition of low SNR is the research focus of this paper.Taking frequency-hopping data link signal as the research object,the recognition method of its modulation mode and the estimation method of its parameters are studied deeply.Aiming at the recognition of modulation mode of frequency hopping data link signal,a modulation mode recognition method based on Intra-Inter Net network is proposed.This method first uses the strong sample fitting ability of CNN and the temporal feature extraction ability of LSTM network to deeply extract the periodic correlation features of data link signals;then it is classified by Softmax;finally,the recognition of seven modulated signals is realized.The simulation results show that when the SNR is in the range of-20 d B to-12 d B,the average recognition accuracy is68.53%;when the signal-to-noise ratio is between-10 d B and-2 d B,the average recognition accuracy is 94.37%;when the SNR is in the range of 0 d B ～ 6 d B,the average recognition accuracy is equal to 99.32%,which effectively improves the modulation recognition accuracy at low SNR.Aiming at the estimation of the parameters of frequency hopping data chain signal,a time-frequency cluster estimation method based on GA optimization is proposed.The method firstly uses genetic algorithm to extract the time-frequency interval of STFT time-frequency graph of data link signal;then the time-frequency ridge is extracted;finally,K-means clustering algorithm was used to classify the time-frequency graph with clustering number of five,and the frequency hopping period and hopping speed were estimated.By classifying the time-frequency ridge with clustering number of six,the hopping frequency is estimated.The simulation results show that the proposed method can accurately estimate the jump period,jump speed and jump frequency of SNR(28)-18 d B,and effectively improve the accuracy of parameter estimation at low SNR compared with the time-frequency ridge parameter estimation method.