Efficient DOA Estimation Based on Coprime Array Interpolation With Deep Unfolding Network

Zhuoqian Jiang; Jingmin Xin; Weiliang Zuo; Nanning Zheng; Akira Sano

doi:10.1109/LSP.2026.3698380

Efficient DOA Estimation Based on Coprime Array Interpolation With Deep Unfolding Network

Zhuoqian Jiang
, Jingmin Xin
, Weiliang Zuo
, Nanning Zheng
, Akira Sano

College of Artificial Intelligence and Robotics

Xi'an Jiaotong University
Keio University

Research output: Contribution to journal › Article › peer-review

Abstract

Coprime arrays increase the degrees of freedom for direction of arrival (DOA) estimation, but virtual-array gaps require costly filling procedures. In this letter, a new DOA estimation method is proposed for coprime arrays based on interpolation and a deep unfolding network. The reconstruction of the interpolated virtual array covariance matrix is formulated as a rank minimization problem and solved using an ADMM-based deep unfolding network with stage-wise learnable parameters and an unsupervised loss inspired by ADMM convergence criteria. Finally, root-MUSIC is employed for DOA estimation. Simulations demonstrate the effectiveness of the proposed method in terms of both computational efficiency and estimation performance.

Original language	English
Journal	IEEE Signal Processing Letters
DOIs	https://doi.org/10.1109/LSP.2026.3698380
State	Accepted/In press - 2026

Keywords

Coprime array
deep unfolding network
direction of arrival (DOA)
interpolation

Access to Document

10.1109/LSP.2026.3698380

Cite this

@article{1753c6fa442c47fe9897930723d7bb12,

title = "Efficient DOA Estimation Based on Coprime Array Interpolation With Deep Unfolding Network",

abstract = "Coprime arrays increase the degrees of freedom for direction of arrival (DOA) estimation, but virtual-array gaps require costly filling procedures. In this letter, a new DOA estimation method is proposed for coprime arrays based on interpolation and a deep unfolding network. The reconstruction of the interpolated virtual array covariance matrix is formulated as a rank minimization problem and solved using an ADMM-based deep unfolding network with stage-wise learnable parameters and an unsupervised loss inspired by ADMM convergence criteria. Finally, root-MUSIC is employed for DOA estimation. Simulations demonstrate the effectiveness of the proposed method in terms of both computational efficiency and estimation performance.",

keywords = "Coprime array, deep unfolding network, direction of arrival (DOA), interpolation",

author = "Zhuoqian Jiang and Jingmin Xin and Weiliang Zuo and Nanning Zheng and Akira Sano",

note = "Publisher Copyright: {\textcopyright} 1994-2012 IEEE.",

year = "2026",

doi = "10.1109/LSP.2026.3698380",

language = "英语",

journal = "IEEE Signal Processing Letters",

issn = "1070-9908",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Efficient DOA Estimation Based on Coprime Array Interpolation With Deep Unfolding Network

AU - Jiang, Zhuoqian

AU - Xin, Jingmin

AU - Zuo, Weiliang

AU - Zheng, Nanning

AU - Sano, Akira

PY - 2026

Y1 - 2026

N2 - Coprime arrays increase the degrees of freedom for direction of arrival (DOA) estimation, but virtual-array gaps require costly filling procedures. In this letter, a new DOA estimation method is proposed for coprime arrays based on interpolation and a deep unfolding network. The reconstruction of the interpolated virtual array covariance matrix is formulated as a rank minimization problem and solved using an ADMM-based deep unfolding network with stage-wise learnable parameters and an unsupervised loss inspired by ADMM convergence criteria. Finally, root-MUSIC is employed for DOA estimation. Simulations demonstrate the effectiveness of the proposed method in terms of both computational efficiency and estimation performance.

AB - Coprime arrays increase the degrees of freedom for direction of arrival (DOA) estimation, but virtual-array gaps require costly filling procedures. In this letter, a new DOA estimation method is proposed for coprime arrays based on interpolation and a deep unfolding network. The reconstruction of the interpolated virtual array covariance matrix is formulated as a rank minimization problem and solved using an ADMM-based deep unfolding network with stage-wise learnable parameters and an unsupervised loss inspired by ADMM convergence criteria. Finally, root-MUSIC is employed for DOA estimation. Simulations demonstrate the effectiveness of the proposed method in terms of both computational efficiency and estimation performance.

KW - Coprime array

KW - deep unfolding network

KW - direction of arrival (DOA)

KW - interpolation

UR - https://www.scopus.com/pages/publications/105040392953

U2 - 10.1109/LSP.2026.3698380

DO - 10.1109/LSP.2026.3698380

M3 - 文章

AN - SCOPUS:105040392953

SN - 1070-9908

JO - IEEE Signal Processing Letters

JF - IEEE Signal Processing Letters

ER -

Efficient DOA Estimation Based on Coprime Array Interpolation With Deep Unfolding Network

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this