RFI抑制技术在射电天文中的应用  

Application of RFI Mitigation Technology in Radio Astronomy

作　　者：张海龙[1,2,3,4] 张亚州 王杰 冶鑫晨[1,4] 王万琼 李嘉 张萌[1,2] 杜旭[1,2] Zhang Hailong;Zhang Yazhou;Wang Jie;Ye Xinchen;Wang Wanqiong;Li Jia;Zhang Meng;Du Xu(Xinjiang Astronomical Observatory,Chinese Academy of Sciences,Urumqi 830011,China;University of Chinese Academy of Sciences,Beijing 100049,China;Key Laboratory of Radio Astronomy,Chinese Academy of Sciences,Nanjing 210033,China;National Astronomical Data Center,Beijing 100101,China)

机构地区：[1]中国科学院新疆天文台,新疆乌鲁木齐830011 [2]中国科学院大学,北京100049 [3]中国科学院射电天文重点实验室,江苏南京210033 [4]国家天文科学数据中心,北京100101

出　　处：《天文研究与技术》2022年第6期613-625,共13页Astronomical Research & Technology

基　　金：国家自然科学基金(12173077,11873082,11803080,12003062);新疆维吾尔自治区天山创新团队计划(2022D14020);中国科学院青年创新促进会;中国科学院天文台站设备更新及重大仪器设备运行专项经费资助;国家重点研发计划(2018YFA0404704,2021YFC2203502);国家天文科学数据中心。

摘　　要：针对射电天文观测过程中的射频干扰(Radio Frequency Interference, RFI)问题,详细分析了国内外台站射频干扰抑制策略。根据各天文台站实际观测过程中遇到的射频干扰问题,分别从主动预防阶段、预相关阶段、后相关阶段、机器学习和深度学习等方面研究了射频干扰的预防策略和抑制方法。详细分析了主动预防阶段可采取的方法,预相关阶段的自适应滤波和空间滤波方法,后相关阶段的VarThreshold, SumThreshold和奇异值分解等方法。探讨了基于机器学习的主成分分析、支持向量机、全卷积神经网络、卷积神经网络、U-Net等相关技术和方法在射频干扰信号处理方面的应用。Aiming at the problem of Radio Frequency Interference(RFI) in the process of radio astronomy observation, the RFI mitigation strategies of domestic and foreign stations are analyzed in detail. According to the RFI problems encountered in the actual observation process of each observatory, the prevention strategies and mitigation methods of RFI are studied from the aspects of active prevention stage, pre-correlation stage, post-correlation stage, machine learning and deep learning. The methods that can be adopted in the active prevention stage, adaptive filtering in the pre-correlation stage, spatial filtering method, VarThreshold, SumThreshold and singular value decomposition in the post correlation stage are systematically analyzed. The applications of principal component analysis based on machine learning, support vector machine, full convolution neural network, convolution neural network, U-Net and other related technologies and methods in RFI signal processing are discussed.

关 键 词：射频干扰 滤波 阈值法 机器学习 

分 类 号：P161[天文地球—天文学]