Gain and phase errors active calibration method based on neural network for arrays with arbitrary geometry

作     者：Han Ziwen Zhang Zhi Guo Yu Han Ziwen;Zhang Zhi;Guo Yu

作者机构：State Key Laboratory of Networking and Switching TechnologyBeijing University of Posts and TelecommunicationsBeijing 100876China 

出 版 物：《The Journal of China Universities of Posts and Telecommunications》 (中国邮电高校学报（英文版）)

年 卷 期：2023年第30卷第2期

页      面：8-17页

核心收录：

学科分类：0711[理学-系统科学] 12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 07[理学] 081104[工学-模式识别与智能系统] 08[工学] 080401[工学-精密仪器及机械] 0804[工学-仪器科学与技术] 080402[工学-测试计量技术及仪器] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by the Key R&D Program of Shandong Province(2020CXGC010109) the Beijing Municipal Science and Technology Project(Z181100003218015) 

主　　题：active array calibration cascaded neural network direction of arrival(DOA)estimation 

摘      要：The manifold matrix of the received signals can be destroyed when the array is with the gain and phase errors,which will affect the performance of the traditional direction of arrival(DOA)estimation *** this paper,a novel active array calibration method for the gain and phase errors based on a cascaded neural network(GPECNN)was *** cascaded neural network contains two parts:signal-to-noise ratio(SNR)classification network and two sets of error estimation *** calibration subnetworks are activated according to the output of the SNR classification network,each of which consists of a gain error estimation network(GEEN)and a phase error estimation network(PEEN),*** disadvantage of neural network topology architecture is changing when the number of array elements varies is addressed by the proposed group calibration ***,due to the data characteristics of the input vector,the cascaded neural network can be applied to arrays with arbitrary geometry without repetitive *** results demonstrate that the GPECNN not only achieves a better balance between calibration performance and calibration complexity than other methods but also can be applied to arrays with different numbers of sensors or different shapes without repetitive training.