Ultra-wideband MIMO Radar Three-dimensional Enhanced Imaging Method

Multiple Input Multiple Output(MIMO)radar imaging technology has a spatial distribution for its array,it can be used to achieve an array of equivalent dense array by virtual aperture feature.It can get the imaging performance of the dense array when reduce the number of elements the complexity of the imaging system.The combination of an ultra-wideband signal with a two-dimensional MIMO array can further obtain the three-dimensional imaging results of the detection area,the more target space distribution information can be obtained.For human targets,accurate measurement for vital signs monitoring and imaging can be provided by ultra-wideband radar signal,and the human posture can be distinguished by radar three-dimensional imaging result.The imaging performance of MIMO radar is limited by hardware conditions and application requirements,the imaging results can be affected directly or indirectly by the noise,clutter and array layout.To obtain high resolution imaging result,the human imaging result is acquired by radar and the enhanced imaging algorithm is developed.The content includes Ultra-wideband(UWB)MIMO radar echo model,human ellipsoid Radar Cross Section(RCS)model,UWB MIMO radar imaging,traditional radar enhanced imaging method based on UWB MIMO radar deconvolution and regularization enhanced imaging method.Firstly,the typical UWB radar waveform and the array characteristics of MIMO radar are researched,the radar echo signal model is established,the one-dimensional range image of echo signal is analyzed,combin with the radar echo model,the real data measurement is analyzed.Obtain the bone nodes by Kinect for the human body as the model parameter,the model is built and the human radar echo is simulated.Then,the impact of the layout mode for MIMO radar array is analyzed,a kind of uniform sampling "mouth" antenna array is designed.The Back projection(BP)imaging algorithm and the Coherence Factor(CF)weighted algorithm is verified by the simulation.Then the performance indexes of target space spectrum and imaging resolution of MIMO radar are analyzed.A radar imaging degeneration model and the Point Spread Function(PSF)model were established.The basic concept of inverse problem is introduced,and the advantages and disadvantages of traditional enhanced imaging method based on deconvolution are introduced,the Lucy-Richardson algorithm is inferred and the point target is simulated.The simulation results prove that the deconvolution method can improve the resolution for point target.The lucy-Richardson algorithm is a more appropriate choice for enhance the human radar image than the CF weighted when don't over increase the range difference between strong and weak targets.However,as the number of iterations increases,the ill-posed problem will be generated and the imaging results will be distorted.Lastly,To solve the inverse problem with deconvolution,the basic principle of regularization is explained,and the advantages and disadvantages of typical regularization methods and their parameter selection are derived and analyzed.Two UWB MIMO radar enhanced imaging methods based on regularization are proposed,and the effectiveness of the proposed algorithm is evaluated by combining the visual identifiability and quantitative index of the imaging results for human body.After comparing the simulation experiment with the real data measurement experiment for the same category methods,the proposed method transforms an ill-posed problem to a wellposed result and choose the optimal decision.The proposed method can improve the image signal-to-noise ratio,have faster convergence speed,lose less image information,have higher image structure similarity with the original imaging results,and have smaller relative error under different noise levels by the simulation experiment and real data measurement experiment.