Millimeter Wave Near-field Three-dimensional Radar Imaging Technology For Concealed Target

Millimeter wave near-field imaging has been playing an important role in the security check area such as personnel concealed target detection,since millimeter waves have the ability to penetrate and with low ionizing radiation.Millimeter wave near-field imaging systems have been applied at airports and other secure locations,which have important value for theory research and commercial application.Millimeter wave imaging systems can be divided into passive imaging system and active imaging system.The active imaging system has becoming more popular than the passive imaging system when used for near-field concealed target imaging,because it has a strong imaging ability and is less affected by the environment.To obtain more information of the target and to improve the following target detection and recognition ability,research on near-field three-dimensional imaging has been paid more and more interest.Based on the background of near-field concealed target imaging and aimed at improving the capability of near-field three-dimensional imaging,this dissertation studies the system design for millimeter wave near-field imaging and sets up a near-field imaging experimental apparatus.Based on the experimental apparatus,theory and methodology for millimeter wave near-field three-dimensional imaging are researched,including the wavenumber domain near-field imaging;the spatial domain near-field imaging;the perturbation based near-field imaging and the sparse based near-field imaging.Chapter 1 introduces the research background and significance of this dissertation and surveys the state-of-the-art development of the millimeter wave near-field three-dimensional imaging.Then it points out some problems required to be further researched and outlines the main work and the organization of this dissertation.Chapter 2 studies the basic theory of near-field three-dimensional imaging,analyzes the parameter design of the near-field imaging system and describes the plane antenna array based millimeter wave near-field three-dimensional imaging experimental apparatus in detail,which can provide effective real measured data for near-field imaging research.Then the traditional interpolation and Fourier transform based near-field three-dimensional imaging algorithm are described and verified effectively by the real measured data.This chapter provides the basis of theory and data for the following study.Chapter 3 studies the millimeter wave near-field three-dimensional imaging method in wavenumber domain.It firstly analyzes the nonuniform characteristics of the near-field three-dimensional data in wavenumber domain and the effect to the imaging procedure.To improve the imaging ability of the nonuniform three-dimensional data,nonuniform fast Fourier transform(NUFFT)is introduced and an imaging method based on fast Gaussian gridding NUFFT(FGG-NUFFT)is proposed.The proposed method is compared with the traditional interpolation imaging method by simulation data and real measured data.It is demonstrated that the proposed method is comparable in both efficiency and accuracy in the full sampled case;and can maintain imaging performance even with the down-sampled measurement while the traditional method has worse results.It can be concluded that the proposed FGG-NUFFT based imaging method is a good choice especially for decreasing the three-dimensional sampled data,which is important for real-time application.Chapter 4 studies the millimeter wave near-field three-dimensional imaging method in spatial domain.Processing imaging in spatial domain can avoid the nonuniform data problem occurred in wavenumber domain.Firstly the direct three-dimensional integral imaging equation is deduced;and then for the aim of decreasing the imaging computational burden,a down-dimensional integral near-field three-dimensional imaging method is proposed.This approache is the tradeoff between the imaging accuracy and the computational efficiency,and can be alternatives for near-field three-dimensional imaging.This chapter enriches the theory of near-field imaging in spatial domain and provides choices for near-field imaging algorithms.Chapter 5 studies the millimeter wave near-field imaging technologies based on perturbation.The near-field imaging problem is often regarded as a summation problem and focuses on dealing with nonuniform spatial spectrum summation.Kinds of approaches are proposed to balance the imaging accuracy and efficiency.Here,from the aspect of perturbation,the imaging procedure is considered as a data perturbation problem.By utilizing first order Taylor expansion,the imaging procedure is decomposed to sum of several Fourier transforms,which is a new imaging view.The imaging form is simple;interpolation is avoided and realization is efficient by directly using Fourier transforms.Simulation data and real measured data have been tested to verify the effectiveness of the proposed method.The perturbation based near-field imaging method can be an alternative for researching near-field imaging,which enriches the millimeter wave near-field three-dimensional imaging.Chapter 6 studies sparse based millimeter wave near-field three-dimensional imaging.To alleviate data collecting burden in near-field three-dimensional imaging,compressed sensing(CS)methodology is introduced.Preliminary investigation of CS based near-field imaging is done and two approaches are proposed in this chapter,which are the interpolation operator based CS imaging method and the perturbation based CS imaging method.Simulation data and real measured data are utilized to verify these two methods.It is demonstrated that both these two CS imaging methods can reconstruct target images effectively from few measuring data,which means that the CS theory are available to be used in near-field imaging for the aim of reducing data collection.And the perturbation based CS imaging method is more efficient,which is promising in the application area of sparse based near-field imaging.The last chapter concludes the whole dissertation and lists some ideas for the future work.