Research On Air-to-Air Operation States Analysis And Identification Of Airborne Fire Control Radar

The airborne fire control radar radiates electromagnetic waves with modulation characteristics into space through the antenna,and can change the antenna scanning mode and waveform parameters according to the requirements of combat missions,complete the search and tracking tasks,and provide target data for airborne weapon guidance.Therefore,quickly judging the air-to-air operation states of the enemy’s airborne fire control radar based on passive reconnaissance signals is the basis for our own combat aircraft to effectively evade threats and carry out counterattacks,which is of great significance.This article is aimed at the wide application of digital array phased array technology and high-speed digital signal processing technology,which leads to the fact that the signal waveform of airborne fire control radar becomes more complicated and the distinction between operation modes is more blurred.It is oriented to the accurate judgment of enemy fighters in electronic warfare.According to the needs of our threatening situation,we will carry out research on air-to-air operation states recognition technology for airborne fire control radars.The main work of the thesis is as follows:1.Through the simulation modeling of the airborne fire control radar,the antenna scanning characteristics and waveform parameter characteristics of different air-to-air functions are summarized,and the difficulty and limitations of the work mode reverse estimation are demonstrated,and then the "air-to-air operation states of airborne fire control radar" is proposed.Through the analysis of radar countermeasures and reconnaissance models,a signal waveform simulator was developed to provide a simulation data set and a theoretical basis for subsequent air-to-air operation states recognition.2.Aiming at the type of data output by the RWR system,an air-to-air operation states recognition algorithm based on DS evidence theory is proposed.The radar feature parameter set involved in the recognition and classification was selected,and the DS evidence theory recognition framework including single feature parameter recognition,multi-feature parameter fusion recognition,multi-period fusion recognition and supplementary judgment was built,and the DS evidence theory recognition network was researched and determined.The basic probability distribution method of the simulation experiment verifies the effectiveness of the recognition algorithm.3.Aiming at the low signal-to-noise ratio full pulse data output by the ESM/ELINT system,a SVM-based air-to-air operation states recognition algorithm is proposed.According to the full pulse data,the radar antenna scanning characteristics and signal waveform parameters of different operation states are analyzed,and the full pulse characteristic parameters such as signal amplitude dispersion and smoothness are proposed.After the network is identified by SVM,the effective classification of the air-to-air operation states is realized.Simulation experiments show that the algorithm can effectively deal with low signal-to-noise ratio full-pulse data,and has a higher recognition accuracy.4.Aiming at the full pulse data of low signal-to-noise ratio and high error and missing pulse ratio output by ESM/ELINT system in complex electromagnetic environment,an air-to-air operation states recognition algorithm based on one-dimensional convolutional neural network is proposed.The one-dimensional convolutional neural network recognition framework can directly preprocess the full-pulse data such as length consistency and data normalization,and then realize the accurate recognition of different air-to-air operation states.The simulation experiment results show that the algorithm does not need to repair the full pulse data or extract the full pulse feature parameters artificially.It eliminates the uncertainty of human judgment as much as possible,and has high recognition accuracy and practical value.