Truncated nuclear norm matrix recover algorithm for direction-of-arrival estimation

The received signal may suffer from continuous missing data due to reasons such as antenna damage, which affects the performance of the direction-of-arrival (DOA) estimation. In order to achieve a better DOA estimation performance, this paper presents a matrix recovery DOA estimation method based on...

Full description

Saved in:
Bibliographic Details
Published in:Signal, image and video processing image and video processing, 2024-06, Vol.18 (4), p.3715-3722
Main Authors: Liu, Fulai, Lou, Xinyue, Zhang, Bo, Du, Ruiyan, Sun, Guozhu
Format: Article
Language:English
Subjects:
Algorithms
Computer Imaging
Computer Science
Direction of arrival
Image Processing and Computer Vision
Mathematical models
Missing data
Multimedia Information Systems
Optimization
Original Paper
Parameter estimation
Pattern Recognition and Graphics
Principal components analysis
Recovery
Signal reconstruction
Signal,Image and Speech Processing
Vision
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The received signal may suffer from continuous missing data due to reasons such as antenna damage, which affects the performance of the direction-of-arrival (DOA) estimation. In order to achieve a better DOA estimation performance, this paper presents a matrix recovery DOA estimation method based on the truncated nuclear norm (TrNN) robust principal component analysis (RPCA). Firstly, the received signal recovery and noise reducing problem of a single snapshot is modeled as a low-rank matrix RPCA recovery problem, and the F-norm is used to constrain the noise entries so as to better separate the low-rank received signal from the noise. Secondly, TrNN is introduced into the low-rank matrix RPCA model in order to make better use of the a priori information on the number of signal sources in the received signal. In the process of TrNN optimization, only large singular values are used corresponding to the number of sources for processing, which reduces the computational complexity of the nuclear norm optimization to some extent without losing the target signal information. The proposed algorithm is used to complete the data of all snapshots. Finally, the DOA parameters are estimated by means of subspace theory. Simulation results show that the proposed algorithm achieves better angular parameter estimation performance with associated low-rank matrix recovery at low SNR.
ISSN:1863-1703
1863-1711
DOI:10.1007/s11760-024-03035-1