The Research On Multipath Ghosts Suppression Of Through-wall Radar Under Unknown Wall Parameters

The through-wall radar provides a unique perspective for the positioning and imaging of targets in closed or irregular environments behind a wall.It has been widely used in the fields of earthquake relief,anti-terrorism and other fields,and has become a current research hotspot.However,due to the existence of targets and walls inside a room or in an enclosed structure,except for the target signal,the echo signals received by the receiving antennas also include the multipath components caused by the interaction between the target to wall and wall to wall,which results in false targets or ghosts and greatly affects the follow-up target identification and classification.On the basis of fully understanding the working principle of the through-wall radar and the basic theory of imaging,this article has done the following research work on obtaining indoor imaging of through-wall radar without ghosts:Firstly,the effect of the wall on the electromagnetic wave is analyzed.Based on the complex indoor scene model,the propagation delay of each grid of each multipath is estimated,and the disadvantages of the traditional time delay estimation,which has a huge calculate complexity,are analyzed,and a fast wall compensation method based on refraction angle estimation is introduced to compensate the time delay,so as to avoid the target from being shifted.Secondly,the estimation of wall parameters is investigated.According to the different components of the echo we extracted,the time-delay iterative estimation method,fast autofocus imaging estimation method and image domain filtering estimation method were used to estimate the wall parameters.In the first method,an error objective function is constructed and the objective function is iteratively optimized to obtain the optimal value.The latter two methods evaluate the quality of the images,which corresponding to the iteratively optimized wall parameters,in the image domain.The parameters corresponding to the imaging result with the highest quality evaluation are the optimal parameters.Thirdly,this paper studies the suppression of multipath ghosts in indoor complex scene with compressed sensing.A new method based on impulse radar signal platform frequency domain compression sensing framework of multi-path exploitation is proposed.The aspect dependence feature is employed between the target and multipath ghosts to analyze the relevance of the position of target,ghosts and the sub-array.Finally,the multipath ghosts will be restrained and high-precision reconstruction of the target is achieved by the image fusion method.This method overcomes the shortcomings of high complexity of time-domain compression sensing in pulse signal and need to know the knowledge of the reflecting geometry in advance.On this basis,the joint coprime array and the aspect dependence feature are combined to further improve the resolution of the reconstruction target.Finally,the multipath ghosts suppression technology based on Bayesian compressed sensing framework is studied.In order to make effective use of the group sparse of multipath components,Bayesian compressive sensing theory is applied to multipath sparse models,and a sparse bayesian learning algorithm based on Complex Temporal correlation is proposed.Then multipath measurement sparse coefficients recovery issue is transformed to a block sparse coefficients recovery issue of a single measurement.This algorithm not only inhibits the generation of multipath ghosts,but also reconstructs indoor targets.Simulation results show that this method has higher reconstruct accuracy and computational efficiency.