Covariance matrix estimation with pipeline structure in array signal processing  

作　　者：Chen Ligang Cai Weijie Liu Zhihui Dong Tao Zeng Hao 

机构地区：[1]School of Microelectronics and Communication Engineering,Chongqing University,Chongqing 400044,China [2]Space Star Technology Co.,Ltd.,Beijing 100095,China

出　　处：《The Journal of China Universities of Posts and Telecommunications》2024年第6期88-94,共7页中国邮电高校学报(英文版)

基　　金：supported by the China Academy of Space Technology(CAST)Innovation Fund (2022LX-312-Y-GZ-15)。

摘　　要：Estimating the covariance matrix of the received signal is a vital step in array signal processing. The traditional estimation method is completed generally by offline way after receiving the continuous sampling snapshots, which cannot meet requirements of real-time performance. In this paper, the calculation of covariance matrix is divided into four pipeline operations, which are multiplication, accumulation, subtraction and division, respectively.The new method leverages the parallel computing function of field programmable gate array(FPGA) to calculate the covariance matrix adopting a pipeline estimation method at each time when the array receives a new sampling snapshot. Compared with the traditional method, the new method not only improves the real-time performance, but also reduces the error in calculation caused by the rapid direction of arrival(DOA) change of the interested signal when the target is highly maneuverable.

关 键 词：array signal processing covariance matrix estimation PIPELINE real-time performance SNAPSHOT 

分 类 号：TN9[电子电信—信息与通信工程]