Research On Radar Anti-towed Decoy Jamming Technology Based On Adaptive Joint Estimation

Towed radar decoy jamming is one of the biggest threats to radar-guided missiles.The complex and changeable radar interference environment causes the pattern of radar interference to change,and the target decoy joint estimation filter designed for the interference pattern of dense false targets cannot meet the actual needs.In response to this problem,this paper mainly studies radar active jamming classification based on machine learning,active jamming classification and recognition based on deep learning,and target/bait joint estimation algorithms based on jamming identification and parameter estimation.The specific research content is as follows:The Introduction part aims at the problem of anti-towed decoy interference,summarizes the research status of the towed decoy interference principle,anti-interference technology and interference classification and identification method,and points out the research direction of this article.Chapter 2 analyzed the towed decoy jamming problem of active radar,the jamming system and the jamming principle of the towed decoy jamming are analyzed,and the jamming scene of the towed decoy jamming is modeled to provide adaptive joint estimation for the following platform.Then,by classifying and modeling several interference patterns,and analyzing their interference effects on the target and the distance Doppler spatial distribution,it will pave the way for the subsequent classification and identification of active interference.Chapter 3 proposes a three-dimensional residual network classification and recognition algorithm for the classification of active radar towed decoy jamming.First,perform simulation experiments on the support vector machine method in the traditional interference classification and recognition algorithm to obtain the classification results.Then,through the time-frequency analysis of different styles of active interference,the time-frequency images of the interference are generated in time series,and the interference classification and recognition algorithm based on the three-dimensional residual network is studied,and the three-dimensional residual network classification algorithm is compared with the two-dimensional residual network through simulation experiments.The comparison between the difference network and the support vector machine method verifies the effectiveness of the algorithm and its performance advantage compared to the existing algorithms.Chapter 4 aiming at the instability of the target/decoy joint estimation algorithm in the complex interference environment,the adaptive target/decoy joint estimation algorithm is studied.Based on the interference identification,the modulation parameter estimation and likelihood function modeling of the interference are carried out.Realize the adaptive filtering estimation of the target decoy.Simulation experiments verify the effectiveness of the algorithm.Chapter 5 summarizes the work of the full text and looks forward to the next step.