Article: Embedding prior knowledge about measurement matrix into neural networks for compressed sensing Journal: International Journal of Wireless and Mobile Computing (IJWMC) 2020 Vol.18 No.3 pp.282 - 288 Abstract: Different algorithms have been proposed for Compressed Sensing (CS). One of the most popular frameworks is Orthogonal Matching Pursuit (OMP). And there are many variants of it. Among various versions, a family of algorithms treats the distribution over an original signal as prior knowledge to obtain a training set for the model, and it achieves a good performance. However, there is other trivial prior knowledge about the measurement matrix that has never been used in compressed sensing in previous work. Hence, we propose a new method to embed the prior knowledge about the measurement matrix and distribution over the original signal into the neural networks for CS. In the end, the empirical support shows that the proposed method brings out a significant improvement. Inderscience Publishers - linking academia, business and industry through research

Title: Embedding prior knowledge about measurement matrix into neural networks for compressed sensing

Authors: Meng Wang; Jing Yu; Chuangbai Xiao; Zhenhu Ning; Yang Cao

Addresses: College of Computer Science, Faculty of Information Technology, Beijing University of Technology, No. 100, Pingleyuan, Chaoyang District, Beijing, China ' College of Computer Science, Faculty of Information Technology, Beijing University of Technology, No. 100, Pingleyuan, Chaoyang District, Beijing, China ' College of Computer Science, Faculty of Information Technology, Beijing University of Technology, No. 100, Pingleyuan, Chaoyang District, Beijing, China ' College of Computer Science, Faculty of Information Technology, Beijing University of Technology, No. 100, Pingleyuan, Chaoyang District, Beijing, China ' School of Information, Beijing Wuzi University, No. 1, Fuhe Street, Tongzhou District, Beijing, China

Abstract: Different algorithms have been proposed for Compressed Sensing (CS). One of the most popular frameworks is Orthogonal Matching Pursuit (OMP). And there are many variants of it. Among various versions, a family of algorithms treats the distribution over an original signal as prior knowledge to obtain a training set for the model, and it achieves a good performance. However, there is other trivial prior knowledge about the measurement matrix that has never been used in compressed sensing in previous work. Hence, we propose a new method to embed the prior knowledge about the measurement matrix and distribution over the original signal into the neural networks for CS. In the end, the empirical support shows that the proposed method brings out a significant improvement.

Keywords: signal processing; compressed sensing; neural networks; measurement matrix.

DOI: 10.1504/IJWMC.2020.106777

International Journal of Wireless and Mobile Computing, 2020 Vol.18 No.3, pp.282 - 288

Accepted: 30 Dec 2019
Published online: 20 Apr 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Embedding prior knowledge about measurement matrix into neural networks for compressed sensing

Keep up-to-date