Font Size: a A A

Cognitive Radar Waveform Parameter Selection And Resource Scheduling For Cooperative Tracking Optimization

Posted on:2022-10-23Degree:MasterType:Thesis
Country:ChinaCandidate:P Z WangFull Text:PDF
GTID:2518306572451934Subject:Information and Communication Engineering
Abstract/Summary:PDF Full Text Request
In modern warfare,higher requirements have been put forward for radars to complete tasks such as detection,tracking,and identification.The electromagnetic environment of the battlefield is complex and changeable,electronic countermeasures technology continues to develop,and the radar combat environment has become more severe.Traditional radar detection technologies and systems cannot meet the needs of high-performance detection,which promotes the development of radar in the direction of intelligence and networking.This paper is oriented to the optimization of tracking performance of single/multi-aircraft/ship-based platform cognitive radar maneuvering targets under the background of clutter and interference,and conducts research on waveform parameter selection and resource scheduling technology.The cognitive waveform parameter selection and cognitive cooperative detection structure are constructed,and the cognitive waveform parameter selection method and resource scheduling method based on heuristic algorithm and reinforcement learning algorithm are proposed.On the basis of traditional reinforcement learning algorithms,a parameter selection method based on the deep double-Q network algorithm and a task allocation and resource scheduling method based on a hierarchical abstract machine algorithm are proposed,which improves the background of airborne/shipborne radars in clutter and interference.Under the maneuvering target tracking ability.First,the basic principles of cognitive radar waveform parameter selection are studied,the basic framework of cognitive waveform parameter selection is constructed,the influence of waveform parameters on target detection and tracking performance is analyzed,and the criterion function for target tracking is analyzed and studied.The optimal waveform parameters under the minimum mean square error criterion,maximum mutual information criterion and minimum wave gate criterion are designed,and the waveform library and reward function of cognitive radar are designed.Secondly,aiming at the parameter selection problem of airborne radar tracking maneuvering target in clutter environment,a simulation model of maneuvering target tracking in clutter environment of airborne radar is constructed,and a heuristic waveform parameter scheduling algorithm based on criterion function is proposed.Further,in order to improve the efficiency of waveform parameter selection,a cognitive waveform parameter selection method based on reinforcement learning is proposed,which uses Q learning,deep Q network algorithm and deep double Q network algorithm to improve tracking performance and effectively improve decision-making.effectiveness.Finally,new scenes with different target motion states and different clutter distributions are constructed to test the applicability of reinforcement learning.Finally,aiming at the problem of coordinated tracking and optimization of netted radars on maneuvering targets under the background of clutter interference,a heuristic algorithm-based cognitive waveform parameter selection method and a hierarchical reinforcement learning-based resource scheduling method are respectively proposed.Aiming at the optimization of the tracking performance of networked radars for air-sea cooperative detection,a heuristic algorithm is proposed for cognitive parameter scheduling;for the problem of target tracking performance degradation under strong/weak interference,a power resource scheduling method based on reinforcement learning is studied;for multiple radars Task allocation and resource management issues in the process of multi-target detection,a hierarchical abstract machine algorithm is proposed,which realizes the simultaneous decision-making of different levels of radar task allocation and resource scheduling.On the basis of meeting the detection requirements of each target,it improves The resource utilization rate of the networked radar.
Keywords/Search Tags:cognitive radar, collaborative detection, maneuvering target tracking, waveform parameter optimization, radar resource management, reinforcement learning
PDF Full Text Request
Related items