Research On Synthetic Aperture Imaging Method Of Through-wall Radar Based On Stolt Migration

Through-wall synthetic aperture radar(SAR)detects targets on the back surface of the wall by transmitting a narrow pulse signal with higher frequency.This paper combines the wave equation of electromagnetic waves to study the stolt migration imaging algorithm of throughwall radar.The main research contents of this paper are as follows:(1)The basic composition and characteristics of Impulse Radar and the factors that affect its range resolution are introduced,and a through-the-wall radar(TWR)model that generates echo data is established.Then combined with the explosion model,the traditional stolt migration algorithm is derived in detail,and the specific calculation formula of the propagation factors due to the different media are given.Finally,the echo data is used to realize the imaging of the traditional stolt migration algorithm.(2)Aiming at the sparseness of the detection scene,a stolt migration imaging algorithm based on Compressed Sensing(CS-stolt)is proposed.This method uses the mathematical model of CS theory to establish a matrix vector model between echo data and imaging data.Due to the sparseness of the detection scene,a smaller amount of data can be completed the reconstruction of the target.The simulation results show that compared with the traditional stolt algorithm,CS-stolt can effectively complete the image reconstruction using only 60%of the data and the reconstruction result has a higher target clutter ratio.(3)Aiming at the phenomenon that the calculation time of CS-stolt migration algorithm based on matrix vector model is longer,CS-stolt migration algorithm based on approximate observation is proposed.The algorithm is realized by constructing the imaging operator and approximate observation operator of the stolt migration imaging algorithm,so that it can directly use the two-dimensional echo data matrix for calculation.Finally,a nonlinear conjugate gradient algorithm is used to achieve target sparse imaging reconstruction.The simulation experiment results show that the CS-stolt imaging algorithm based on approximate observation can complete the image reconstruction well using only 40%of the data volume.Compared with the CS-stolt migration imaging algorithm based on the matrix vector model,the calculation time is reduced by an average of 100 times.