Research On Waveform Parameter Selection Algorithm Of Intelligent Radar

As a long-range target detection equipment,radar plays an extremely important role in the military field.In order to improve the adaptive ability of radar system in complex time-varying environment,intelligent radar forms a closed-loop interaction system between radar and environment by constructing the information feedback mechanism from receiver to transmitter and combining the artificial intelligence technology driven by knowledge and data,which can effectively improve the system performance and adaptive ability of radar to new environment,new targets and new tasks,the development of intelligent radar has become the research hotspot of the next generation radar.This thesis aims to study the waveform parameter selection of intelligent radar in target tracking scenarios.The specific work of this thesis is as follows:1.The performance bottleneck of traditional radar in complex combat environment is discussed.The principle framework of intelligent radar in target tracking scenario is given and the corresponding mathematical model is constructed.By establishing an intelligent radar signal model,the influence of transmitting waveform parameters on the covariance of measured noise in the Bayesian filter is analyzed,and on this basis,three kinds of intelligent radar waveform parameter selection criteria functions are given as the basis of the research in the following chapters.2.Aiming at the problem of tracking highly maneuverable targets by a single radar,this thesis proposed an IMM based waveform parameter selection algorithm.In the proposed algorithm,the prediction covariance of each sub-filters in IMM model to target state is weighted and fused together with criterion function to realize the intelligent selection of waveform parameters.Simulation results show that compared with traditional radar,the proposed algorithm can improve the effective probability of correct motion model in IMM model set in both 2D and 3D space,and then improve the tracking performance.In addition,the tracking performance of IMM waveform parameter selection algorithm under three different criterion functions is compared,and the results show that the minimum mean square error criterion has the best performance.3.Aiming at the problem of tracking maneuvering targets in radar sensor network,this thesis proposed a joint waveform parameter selection algorithm based on reinforcement learning.Meanwhile,in order to solve the computational load problem of finding the optimal waveform parameters,a waveform parameter selection machine is constructed by reinforcement learning.The results show that compared with the traditional radar,this algorithm can effectively improve the target tracking performance of radar sensor network with lower computational cost.4.Aiming at the coordination problem of radar tracking performance and resource scheduling,this thesis proposed a fuzzy logic based waveform parameter selection algorithm.In this algorithm,a constrained objective optimization function is presented which can account for both tracking performance and resource scheduling,and the weight of the optimization function is adjusted adaptively by a fuzzy logic system.The simulation results show that the algorithm can effectively reduce the radar resource usage on the premise of satisfying the preset radar performance requirements.In summary,this thesis proposes three waveform parameter selection algorithms for both single radar and radar network in target tracking scenarios,all of which effectively improve the tracking performance of the radar system,and provide a theoretical basis for the selection of waveform parameters of intelligent radar.