Research On Multi-path SAR Image Suppression Based On RPCA

Ground Based Synthetic Aperture Radar(GBSAR),as a new high-precision and large field of view measurement tool,has the ability to provide high-resolution non-contact imaging all day,all weather,and has a wide range of applications in monitoring areas such as mining areas,slopes,and dams.When facing the demand for long-term fixed point continuous monitoring,GBSAR usually chooses to build indoor enclosed monitoring stations to avoid the impact of harsh outdoor environments on equipment.The radar is placed on the operating platform,and the antenna sends and receives signals through a transparent cover.However,compared to outdoor environments,the echo signals received by radar in closed environments are susceptible to interference from close range strong scattering signals caused by multipath in the surrounding space.After subsequent imaging processing,the echo signal obtained by the radar receiver in a closed space will exhibit bright lines distributed along the range direction in close range multipath strong scattering,seriously affecting the imaging quality.Due to the traditional SAR image interference suppression methods,such as azimuth signal averaging and principal component analysis(PCA)suppression of the original echo signal,they cannot effectively eliminate close range multipath strong scattering interference.Therefore,this article proposes Robust Principal Component Analysis(RPCA)to address this issue,mainly focusing on the range Doppler domain analysis of linear scanning ground-based SAR models,the range Doppler domain model analysis of close range strong scattering signals caused by multipath signals,and the suppression of RPCA algorithm.The research content of this article is as follows:(1)Analyzed the sources of close range multipath strong scattering interference in linear scanning ground-based SAR in closed space scenarios,analyzed the sources of direct wave coupling interference in closed space scenarios,analyzed the distribution characteristics of multipath interference in the azimuth direction,established a range Doppler domain model for direct wave coupling signal and close range strong scattering signal caused by multipath,and verified the effectiveness of the model through simulation.(2)Analyzed the principles of traditional PCA algorithm and RPCA algorithm in this article,and analyzed the specific process steps for suppressing interference signals based on PCA and RPCA in close range multipath strong scattering and direct wave coupling.PCA achieves clutter suppression by performing SVD decomposition on the echo,extracting and canceling multipath strong scattering signals.The RPCA algorithm decomposes the echo signal into low rank and sparse parts in the range Doppler domain,utilizing the low rank characteristics of multipath strong scattering signals in the range Doppler domain and the sparse characteristics of scene signals to effectively separate interference signals from scene signals.This paper also proposes an optimal selection method of RPCA regularization coefficients based on correlation analysis to achieve better separation of interference and scene.(3)Perform algorithm processing on experimental data collection,and compare the multipath strong scattering suppression results with PCA suppression results after selecting the optimal threshold for RPCA.Subsequently,it was proposed to use image entropy as an indicator to validate the RPCA based multipath strong scattering signal suppression method,which can better suppress multipath strong scattering signals while preserving the scene echo signal.