Radar Individual And Working Status Recognition

With the continuous advancement of electronic information technology,the battlefield of modern war has expanded from the traditional land,sea,and air battlefields to network and electronic space.Electronic countermeasures have become an indispensable part of modern warfare.Radar reconnaissance system and radar jamming system are the core of electronic countermeasures.By identifying the intercepted radar radiation source signals,the radar’s working mode and individual information are judged,and the analysis is used to evaluate enemy attacks,own defense,and battlefield situation assessment Provide intelligence support.But nowadays,the electromagnetic environment is getting worse and worse.Radar source signals are more variable and there is a lot of interference.Therefore,it is of great significance to perform radar reconnaissance tasks quickly and accurately.Radar jamming is based on reconnaissance,interfering with enemy radar electronics and systems based on the battlefield situation,causing them to lose or reduce their effectiveness.This paper mainly studies the radar individual recognition problem in radar reconnaissance system,the incremental learning problem of radar working mode recognition,and the radar interference decision problem in radar electronic countermeasures.First,a radar radiation individual identification algorithm based on adaptive longest common subsequence(ada-LCSS)is proposed for the radar radiation individual identification problem.This algorithm improves the longest common subsequence matching algorithm to reduce the influence of noise on individual recognition results.This paper introduces the solution of the individual identification problem of the radiation source and the extraction process of the pulse envelope,and proposes an improved ada-LCSS algorithm.This method combines the characteristics of the attenuation and interference of the radar signal during the propagation process to change the original The linear weight is changed to a non-linear weight,and the radar radiation source is identified by the similarity calculation formula.Then,a new data block-based integrated radar working state incremental learning algorithm is proposed for the incremental learning problem in radar source identification.This algorithm improves the concept drift processing method(DTEL)based on differential model selection and knowledge transfer,and addresses the problem of resource waste caused by updating the integration model every time a new data block is encountered in DTEL.By dding a concept drift detection mechanism Control the timing of the start of incremental learning to shorten the training time,and at the same time use the selection of a new differential model selection strategy to improve the integration model training speed on the premise of ensuring the correct rate of incremental learning classification.Finally,a radar interference decision-making method based on adaptive heuristic Q-Learning is proposed for radar interference decision-making.First of all,starting from the basic problems in radar interference decision analysis,the components of the interference strategy,radar interference decision methods,and interference effect evaluation methods are introduced.The relationship between changes in radar operating conditions and radar interference effectiveness evaluation is studied.Then,an improved radar interference decision-making method is proposed,which guides the choice of interference actions in the reinforcement learning process by adaptively combining the interference trend table(IIT),thereby shortening the learning time and obtaining the best interference strategy.