Waveform Parameters Intelligent Selection Technology For Tracking Optimization Of Radar Cooperative Detection

Radar is responsible for the main tasks of reconnaissance,detection,identification and tracking in the contemporary detection system.The modern battlefield environment is becoming more and more complex,the detection range of a single radar is limited,and it will be restricted by the airspace coverage blind zone and the Doppler blind zone.The diversified forms of operations require that radar systems gradually shift from single radar systems to multi-platform radar systems.This paper is based on a multi-radar cooperative layered detection structure,based on criterion functions,neural networks,and reinforcement learning methods to achieve environmental awareness and radar parameter optimization.The tracking performance of the collaborative detection system is analyzed in different environments.This article first introduces the background significance of the subject,introduces the different theoretical foundations of cooperative detection,and the research and development of cooperative systems at home and abroad;then analyzes the various modules of cooperative detection radar;then builds the launch waveform model of the cooperative detection system In order to ensure the orthogonality of the waveforms between the radars,at the same time analyze the influence of the radar waveform parameters on the signal-to-noise ratio and measurement,and derive the calculation method of the signal-to-noise ratio used in the subsequent simulation experiments of the cooperative detection system.Subsequently,a collaborative detection system based on a single base and fusion center in a clutter-free environment is established;a layered criterion function method for collaborative detection structure is proposed to analyze the effect of adjusting waveform parameters on tracking single target accuracy.Further to the complex environment where clutter exists,a combination of extended Kalman filtering and probabilistic data correlation method is proposed to analyze how to adjust the waveform parameters when the environment changes in complex scenes such as clutter.Aiming at the problem of intelligent selection of waveform parameters in single target tracking,a multi-radar single target cooperative tracking processing framework based on neural network is proposed;under the constraint of dual criterion functions of single radar and hierarchical two-fusion central radar,collaborative detection is realized through neural network learning Waveform parameter selection.In order to improve the training speed and cope with the problem of adaptive parameter selection under scene changes,a waveform parameter selection method based on multi-agent reinforcement learning was carried out,and a waveform parameter selection based on hierarchical reinforcement learning was proposed;simulation experiments verified this The method can realize online waveform parameter selection faster while ensuring tracking performance.