Research On FMCW Radar Based Joint Estimation Algorithm Of Multidimensional Parameters

Frequency Modulated Continuous Wave(FMCW)radar has the advantages of no blind zone in range,high ranging accuracy and simple structure,which makes its application field from military to civilian.With the widespread application of FMCW radar,the accuracy and real-time performance of range,velocity and angle parameter estimation of its targets have been put forward to higher requirements.Multi-dimensional parameter joint estimation algorithm is favored by researchers because it can combine the characteristics of many parameters to improve the accuracy of parameter estimation.The signals collected by the FMCW radar generally contain the three-dimensional parameters of the target's range,speed and angle,and the multi-dimensional parameter joint estimation is to use a certain algorithm to combine the characteristics of the threedimensional parameters to achieve the estimation of multiple dimensional parameters of the target.There are some major characteristics of current multi-dimensional parameter estimation: First,the multi-dimensional Fast Fourier Transform(FFT)algorithm has low accuracy due to the effects of the fence effect and spectrum leakage;Second,In the multitarget scenario,the problem of estimation error is easy to occur because the number of targets is unknown and the interference between targets is serious.Third,Multidimensional subspace decomposition algorithms can be used to achieve high-precision range,velocity and angle estimation,but the algorithm has high complexity and poor practicability.Based on those,this thesis has carried out research on multi-dimensional parameter joint estimation algorithm based on FMCW radar.The main work of this thesis is as follows:Firstly,according to the characteristics of FMCW radar intermediate frequency signals,a two-dimensional FFT algorithm is used to obtain a two-dimensional range velocity spectrum.Then the variable step size iterative interpolation algorithm is used to improve the fence effect in algorithm estimation,and the accuracy of distance and speed estimation of the target is improved.Finally,the feasibility of the algorithm is verified by simulation and experiment.Secondly,when the number of targets is unknown,a two-dimensional combined adaptive constant false alarm detection(CFAR)algorithm is first proposed to achieve target detection in various clutter environments.Then a strong scattering point aggregation algorithm is used to classify point targets with multiple scattering points to achieve the estimation of the number of targets in the signal,and record the position information corresponding to the strong scattering point target.Thirdly,based on the correlation between the three domain parameters in the radar signal,a low-complexity and high-precision three-parameter joint estimation method based on distance matching is proposed.First of all,based on the recorded position information,a variable step size iterative interpolation algorithm is used for the targets in the twodimensional frequency spectrum and the initial range to achieve high-precision distance and speed estimation of each target.Then a combined algorithm of multi-sweep noncoherent accumulation and spatial smoothing multiple signal classification(MUSIC)is used to realize the joint estimation of target range and angle.Finally,the match between the target parameters is achieved by using the initial range information as the hub.