Interference Mitigation for Synthetic Aperture Radar Data using Tensor Representation and Low-Rank Approximation

Mingliang Tao, Jieshuang Li, Jia Su, Yifei Fan, Ling Wang, Zijing Zhang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

7 Scopus citations

Abstract

Radio frequency interference (RFI) is a critical issue to synthetic aperture radar (SAR), which would cause great distortions to amplitude and phase information of the received echoes. Most of the existing literatures deal with the interference separation problem in time domain, frequency domain, or time-frequency domain using the matrix representation and matrix optimization tools, without further exploiting the correlation among multiple dimensional measurements. This paper proposes an interference separation for SAR data using tensor representation by formulating a novel time-frequency azimuth tensor. Then, the low-rank property of the interference is utilized and the interference contribution is estimated using low rank tensor approximation. Experimental results demonstrate that the interference components is effectively extracted, and well imaging results could be recovered.

Original languageEnglish
Title of host publication2020 33rd General Assembly and Scientific Symposium of the International Union of Radio Science, URSI GASS 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9789463968003
DOIs
StatePublished - Aug 2020
Event33rd General Assembly and Scientific Symposium of the International Union of Radio Science, URSI GASS 2020 - Rome, Italy
Duration: 29 Aug 20205 Sep 2020

Publication series

Name2020 33rd General Assembly and Scientific Symposium of the International Union of Radio Science, URSI GASS 2020

Conference

Conference33rd General Assembly and Scientific Symposium of the International Union of Radio Science, URSI GASS 2020
Country/TerritoryItaly
CityRome
Period29/08/205/09/20

Fingerprint

Dive into the research topics of 'Interference Mitigation for Synthetic Aperture Radar Data using Tensor Representation and Low-Rank Approximation'. Together they form a unique fingerprint.

Cite this