Disjunct Matrices for Compressed Sensing

Sasmal, P and Thoota, SS and Murthy, CR (2019) Disjunct Matrices for Compressed Sensing. In: 44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019, 12 - 17 May 2019, Brighton, pp. 4888-4892.

PDF ICASSP_2019.pdf - Published Version Restricted to Registered users only Download (11MB) | Request a copy

Abstract

Disjunct matrices play a central role in non-adaptive group testing, as they provide necessary and sufficient conditions for identifying defective items from a large population using a small number of tests. In this paper, we show that binary disjunct matrices can also be very useful for recovering sparse signals from underdetermined linear measurements. They admit non-iterative, ultra-low complexity recovery of sparse signals. Binary measurement matrices have the added benefit of being friendly for hardware implementation. Further, we generalize the notion of disjunctness to matrices with arbitrary (non-binary) entries and show that such matrices also admit similar fast sparse vector recovery algorithms. We empirically demonstrate that disjunct matrices can recover denser signals than recent non-iterative sparse recovery algorithms. © 2019 IEEE.

Item Type:	Conference Paper
Publication:	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Publisher:	Institute of Electrical and Electronics Engineers Inc.
Additional Information:	The copyright for this article belongs to Institute of Electrical and Electronics Engineers Inc.
Keywords:	Audio signal processing; Compressed sensing; Iterative methods; Matrix algebra; Recovery; Speech communication, Binary measurements; Disjunct matrices; Hardware implementations; Large population; Linear measurements; Non-adaptive group testing; Recovery algorithms; Sparse signal recoveries, Signal reconstruction
Department/Centre:	Division of Electrical Sciences > Electrical Communication Engineering
Date Deposited:	16 Dec 2022 07:49
Last Modified:	16 Dec 2022 07:49
URI:	https://eprints.iisc.ac.in/id/eprint/78378

Actions (login required)

View Item