Research On Range Migration Algorithm Of Active Near-field Millimeter-wave Imaging

In recent years,it becomes more and more important to tackle the growing terrorist threat.Thus,the security imaging system,which can be applied in detection of concealed weapons and explosives such as gun,knife,grenade and combustible materials,carried by passengers,is becoming particularly important.Millimeter-wave is the ideal frequency band for personnel surveillance imaging at airports and other public places,because of its advantages of high-resolution,good penetrability to clothes,and radiation-free.This paper focuses on range migration algorithm for active near-field millimeter-wave imaging,and researches on some imaging problems of MIMO linear array scanning system and single channel circular scanning system.Firstly,range migration algorithm for dense linear array and MIMO linear array are researched.To correct the curvature of the wavefront in the near-field,the raw data is resampled to a uniform grid in spatial frequency domain by the use of interpolation after phase migration.The aliasing and ghost caused by the sparsity of the element in the MIMO array are suppressed.Secondly,based on the theory of Circular Synthetic Aperture Radar,an circular scanning imaging system with squint angle for security surveillance to human body is presented.There are two imaging modes for this system.The imaging area is the circle parallel to the scanning plane of the antenna,or the cylinder perpendicular to that.In this paper,range migration algorithm for two imaging modes are studied,respectively.Due to the polar data acquired under circular scanning,range migration algorithm in polar format is researched under the mode of circular imaging area.The algorithm is adapted to the original polar coordinates data.At first,the raw data in slant plane is converted to data in ground plane.Then,polar spatial frequency domain data is obtained after matched filtering.The final reconstructed image is obtained by the use of polar-to-rectangular interpolation and two-dimensional inverse FFT.The algorithm is easy to implement with less computation,and improves the imaging focus effect due to the absence of approximation in phase.Thirdly,for the imaging mode of cylinder imaging area,the high sidelobe level of reconstructed image is caused by the approximation of phase item of range migration algorithm.The cause of the high-level sidelobe is analyzed by comparing the precise phase's imaging results and the approximate phase's imaging results in this paper.Finally,an experimental ka-band turntable imaging system composed of a vector network analyzer,a high precision turntable,a tripod and a pair of horn antennas is built.For two imaging modes of the system,raw data of gun and scissors is collected and the image is reconstructed using range migration algorithm.The experimental results show that the reconstructed image of circle imaging mode is better than cylinder imaging mode.The former's imaging resolution is up to 1-10 mm,so the object is quite easy to be recognized.Although the latter's imaging has worse resolution and high-level sidelobe,the outline of dangerous objects can be still recognized.The experimental results demonstrate the effectiveness of the proposed system in the detection of concealed weapons carried on human body.