Radio frequency interference mitigation using pseudoinverse learning autoencoders

作     者：Hong-Feng Wang Mao Yuan Qian Yin Ping Guo Wei-Wei Zhu Di Li Si-Bo Feng Hong-Feng Wang;Mao Yuan;Qian Yin;Ping Guo;Wei-Wei Zhu;Di Li;Si-Bo Feng

作者机构：Image Processing and Pattern Recognition LaboratorySchool of Artificial IntelligenceBeijing Normal UniversityBeijing 100875China CAS Key Laboratory of FASTNational Astronomical ObservatoriesChinese Academy of SciencesBeijing 100101China School of Information ManagementDezhou UniversityDezhou 253023China Image Processing and Pattern Recognition LaboratorySchool of System ScienceBeijing Normal UniversityBeijing 100875China Institute for Astronomical ScienceDezhou UniversityDezhou 253023China University of Chinese Academy of SciencesBeijing 100049China Hanvon Technology Co.LtdBeijing 100193China NAOC-UKZN Computational Astrophysics CentreUniversity of KwaZulu-NatalDurban 4000South Africa 

出 版 物：《Research in Astronomy and Astrophysics》 (天文和天体物理学研究（英文版）)

年 卷 期：2020年第20卷第8期

页      面：121-128页

核心收录：

学科分类：0709[理学-地质学] 07[理学] 0708[理学-地球物理学] 070401[理学-天体物理] 0704[理学-天文学] 0825[工学-航空宇航科学与技术] 

基　　金：the National Natural Science Foundation of China(NSFC,Grant Nos.11988101,61472043,11743002,11873067,11690024,11673005 and 11725313) the Outstanding Youth Fund Project of Natural Science Fund of Shandong Province(Grant No.ZR2019YQ03) supported by the Joint Research Fund in Astronomy(U1531242)under cooperative agreement between the NSFC and the Chinese Academy of Sciences(CAS)supported by the Chinese Academy of Science Pioneer Hundred Talents Program the Strategic Priority Research Program of the Chinese Academy of Sciences(Grant No.XDB23000000)。 

主　　题：pulsars:general methods:numerical methods:data analysis 

摘      要：Radio frequency interference(RFI)is an important challenge in radio astronomy.RFI comes from various sources and increasingly impacts astronomical observation as telescopes become more sensitive.In this study,we propose a fast and effective method for removing RFI in pulsar data.We use pseudo-inverse learning to train a single hidden layer auto-encoder(AE).We demonstrate that the AE can quickly learn the RFI signatures and then remove them from fast-sampled spectra,leaving real pulsar signals.This method has the advantage over traditional threshold-based filter method in that it does not completely remove contaminated channels,which could also contain useful astronomical information.