The University of Southampton
University of Southampton Institutional Repository

Distributed multi-view sparse vector recovery

Distributed multi-view sparse vector recovery
Distributed multi-view sparse vector recovery
In this paper, we consider a multi-view compressed sensing problem, where each sensor can only obtain a partial view of the global sparse vector. Here the partial view means that some arbitrary and unknown indices of the global vector are unobservable to that sensor and do not contribute to the measurement outputs. The sensors aim to collaboratively recover the global state vector in a decentralized manner. We formulate
this recovery problem as a bilinear optimization problem relying on a factored joint sparsity model (FJSM), in which the variables are factorized into a node-specific sparse local masking vector and the desired common sparse global vector. We first theoretically analyze the general conditions guaranteeing the global vector’s successful recovery. Then we propose a novel in network algorithm based on the powerful distributed alternating direction method of multipliers (ADMM), which can reconstruct the vectors and achieve consensus among nodes concerning the estimation of the global vector. Specifically, each node alternately updates the common global vector and its local masking vector, and then it transfers the estimated global vector to its neighboring nodes for further updates. To avoid potential divergence of the iterative algorithm, we propose an early stopping rule for the
estimation of the local masking vectors and further conceive an estimation error-mitigation algorithm. The convergence of the proposed algorithms is theoretically proved. Finally, extensive simulations validate their excellent performance both in terms of the convergence and recovery accuracy.
Sensor network, alternating direction method of multipliers (ADMM), distributed compressed sensing, distributed optimization, multi-view sparse vector recovery
1053-587X
1448 - 1463
Tian, Zhuojun
bb049ebe-90f1-4f00-855f-2a096aed47f5
Zhang, Zhaoyang
5951d239-6a4e-41d1-a2e3-033e7696a939
Hanzo, Lajos
66e7266f-3066-4fc0-8391-e000acce71a1
Tian, Zhuojun
bb049ebe-90f1-4f00-855f-2a096aed47f5
Zhang, Zhaoyang
5951d239-6a4e-41d1-a2e3-033e7696a939
Hanzo, Lajos
66e7266f-3066-4fc0-8391-e000acce71a1

Tian, Zhuojun, Zhang, Zhaoyang and Hanzo, Lajos (2023) Distributed multi-view sparse vector recovery. IEEE Transactions on Signal Processing, 71, 1448 - 1463. (doi:10.1109/TSP.2023.3267995).

Record type: Article

Abstract

In this paper, we consider a multi-view compressed sensing problem, where each sensor can only obtain a partial view of the global sparse vector. Here the partial view means that some arbitrary and unknown indices of the global vector are unobservable to that sensor and do not contribute to the measurement outputs. The sensors aim to collaboratively recover the global state vector in a decentralized manner. We formulate
this recovery problem as a bilinear optimization problem relying on a factored joint sparsity model (FJSM), in which the variables are factorized into a node-specific sparse local masking vector and the desired common sparse global vector. We first theoretically analyze the general conditions guaranteeing the global vector’s successful recovery. Then we propose a novel in network algorithm based on the powerful distributed alternating direction method of multipliers (ADMM), which can reconstruct the vectors and achieve consensus among nodes concerning the estimation of the global vector. Specifically, each node alternately updates the common global vector and its local masking vector, and then it transfers the estimated global vector to its neighboring nodes for further updates. To avoid potential divergence of the iterative algorithm, we propose an early stopping rule for the
estimation of the local masking vectors and further conceive an estimation error-mitigation algorithm. The convergence of the proposed algorithms is theoretically proved. Finally, extensive simulations validate their excellent performance both in terms of the convergence and recovery accuracy.

Text
Distributed_Multi_View_Sparse_Vector_Recovery - Author's Original
Available under License Creative Commons Attribution.
Download (1MB)

More information

Accepted/In Press date: 6 April 2023
e-pub ahead of print date: 3 May 2023
Published date: 2023
Additional Information: Funding Information: The work of Zhuojun Tian and Zhaoyang Zhang was supported in part by National Natural Science Foundation of China under Grants U20A20158 and 61725104, in part by the National Key R&D Program of China under Grants 2020YFB1807101 and 2018YFB1801104, and in part by Zhejiang Provincial Key R&D Program under Grant 2023C01021. The work of Lajos Hanzo was supported in part by the Engineering and Physical Sciences Research Council under Projects EP/W016605/1 and EP/X01228X/1, in part by the European Research Council's Advanced Fellow Grant QuantCom under Grant 789028. Publisher Copyright: © 1991-2012 IEEE.
Keywords: Sensor network, alternating direction method of multipliers (ADMM), distributed compressed sensing, distributed optimization, multi-view sparse vector recovery

Identifiers

Local EPrints ID: 476914
URI: http://eprints.soton.ac.uk/id/eprint/476914
ISSN: 1053-587X
PURE UUID: 46af65b3-5e97-4e49-81a9-c73dea786bd3
ORCID for Lajos Hanzo: ORCID iD orcid.org/0000-0002-2636-5214

Catalogue record

Date deposited: 19 May 2023 16:35
Last modified: 18 Mar 2024 02:36

Export record

Altmetrics

Contributors

Author: Zhuojun Tian
Author: Zhaoyang Zhang
Author: Lajos Hanzo ORCID iD

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×