Research On Denoising And Detection Technology Of Low Probability Radar Radiation Source Signal

The electromagnetic environment of modern battlefield is complex and the battlefield situation is ever changing.How to find the enemy,lock and launch missiles ahead of the opposite side,and evacuate the battlefield are important to personnel life of both side.In this background,the radar stealth technology came into being.The signal of low probability of intercept radar is with low detectability,which is not easy to find the enemy radar position and destroy battlefield equipment.The alarm and confrontation technology for low probability of intercept radar is the key to the survival of combat units such as fighters and ships.The low-intercept probability radar warning and confrontation technology is based on the premise of the radar radiation weak signal denoising and detection of the signal.After this,the feature extraction and inversion reconstruction have important practical significance.In this paper,we focus on the noise reduction and detection techniques of low probability of intercept radar.The EMD and wavelet theory are combined to denoise.The time-frequency distribution of the signal is used to classify the low-intercept probability radar radiation source.Firstly,several types of low-intercept probability radar signal are modeled.The research gathers the common low-probability probabilistic radar signal types,including frequency modulation signals and phase coded signals.The signal characteristics and how to realize low detectability are analyzed emphatically.Then,the noise of intercepted low intercept probability radar signal is reduced.I use the time-frequency peak filtering algorithm to a certain extent,to achieve signal enhancement.Because the empirical mode decomposition algorithm can effectively deal with high frequency noise and wavelet algorithm can filter out low frequency noise.The two algorithms are combined to reduce the noise of low intercept probability radar signal.Finally,the detection and processing technology of low detectable radar signal for noncooperative target equipment is studied.In order to realize the identification of radiation source,the energy of the modulated signal is extracted by using the marginal frequency binarization algorithm,After the main component analysis into the multi-layer sensor,in order to achieve low probability of intercept radar signal modulation type classification.In this study,the low-intercept probability signal is processed by timefrequency peak filtering and iterative gap empirical mode decomposition and noise reduction algorithm.Then,the signal parameters of time-frequency analysis are used to detect the signal parameters,and the signal modulation type is detected by multi-layer sensor of neural network.The algorithm can effectively deal with low probability of intercepted signals with SNR as low as-20 d B,the average classification detection accuracy rate of 90% or more.