A data-driven compressive sensing framework tailored for energy-efficient wearable sensing

Kai Xu, Yixing Li, Fengbo Ren

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Compressive sensing (CS) is a promising technology for realizing energy-efficient wireless sensors for long-term health monitoring. However, conventional model-driven CS frameworks suffer from limited compression ratio and reconstruction quality when dealing with physiological signals due to inaccurate models and the overlook of individual variability. In this paper, we propose a data-driven CS framework that can learn signal characteristics and personalized features from any individual recording of physiologic signals to enhance CS performance with a minimized number of measurements. Such improvements are accomplished by a co-training approach that optimizes the sensing matrix and the dictionary towards improved restricted isometry property and signal sparsity, respectively. Experimental results upon ECG signals show that the proposed method, at a compression ratio of 10×, successfully reduces the isometry constant of the trained sensing matrices by 86% against random matrices and improves the overall reconstructed signal-to-noise ratio by 15dB over conventional model-driven approaches.

Original languageEnglish (US)
Title of host publication2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages861-865
Number of pages5
ISBN (Electronic)9781509041176
DOIs
StatePublished - Jun 16 2017
Event2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - New Orleans, United States
Duration: Mar 5 2017Mar 9 2017

Other

Other2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017
CountryUnited States
CityNew Orleans
Period3/5/173/9/17

Fingerprint

Glossaries
Electrocardiography
Signal to noise ratio
Health
Monitoring
Sensors

Keywords

  • Data-driven compressive sensing
  • Internet of things (IoT)
  • mobile healthcare
  • wearable sensing

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Xu, K., Li, Y., & Ren, F. (2017). A data-driven compressive sensing framework tailored for energy-efficient wearable sensing. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings (pp. 861-865). [7952278] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2017.7952278

A data-driven compressive sensing framework tailored for energy-efficient wearable sensing. / Xu, Kai; Li, Yixing; Ren, Fengbo.

2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 861-865 7952278.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Xu, K, Li, Y & Ren, F 2017, A data-driven compressive sensing framework tailored for energy-efficient wearable sensing. in 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings., 7952278, Institute of Electrical and Electronics Engineers Inc., pp. 861-865, 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017, New Orleans, United States, 3/5/17. https://doi.org/10.1109/ICASSP.2017.7952278
Xu K, Li Y, Ren F. A data-driven compressive sensing framework tailored for energy-efficient wearable sensing. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. p. 861-865. 7952278 https://doi.org/10.1109/ICASSP.2017.7952278
Xu, Kai ; Li, Yixing ; Ren, Fengbo. / A data-driven compressive sensing framework tailored for energy-efficient wearable sensing. 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 861-865
@inproceedings{77e6a4ce5a82449999239585fd8a98d1,
title = "A data-driven compressive sensing framework tailored for energy-efficient wearable sensing",
abstract = "Compressive sensing (CS) is a promising technology for realizing energy-efficient wireless sensors for long-term health monitoring. However, conventional model-driven CS frameworks suffer from limited compression ratio and reconstruction quality when dealing with physiological signals due to inaccurate models and the overlook of individual variability. In this paper, we propose a data-driven CS framework that can learn signal characteristics and personalized features from any individual recording of physiologic signals to enhance CS performance with a minimized number of measurements. Such improvements are accomplished by a co-training approach that optimizes the sensing matrix and the dictionary towards improved restricted isometry property and signal sparsity, respectively. Experimental results upon ECG signals show that the proposed method, at a compression ratio of 10×, successfully reduces the isometry constant of the trained sensing matrices by 86{\%} against random matrices and improves the overall reconstructed signal-to-noise ratio by 15dB over conventional model-driven approaches.",
keywords = "Data-driven compressive sensing, Internet of things (IoT), mobile healthcare, wearable sensing",
author = "Kai Xu and Yixing Li and Fengbo Ren",
year = "2017",
month = "6",
day = "16",
doi = "10.1109/ICASSP.2017.7952278",
language = "English (US)",
pages = "861--865",
booktitle = "2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - A data-driven compressive sensing framework tailored for energy-efficient wearable sensing

AU - Xu, Kai

AU - Li, Yixing

AU - Ren, Fengbo

PY - 2017/6/16

Y1 - 2017/6/16

N2 - Compressive sensing (CS) is a promising technology for realizing energy-efficient wireless sensors for long-term health monitoring. However, conventional model-driven CS frameworks suffer from limited compression ratio and reconstruction quality when dealing with physiological signals due to inaccurate models and the overlook of individual variability. In this paper, we propose a data-driven CS framework that can learn signal characteristics and personalized features from any individual recording of physiologic signals to enhance CS performance with a minimized number of measurements. Such improvements are accomplished by a co-training approach that optimizes the sensing matrix and the dictionary towards improved restricted isometry property and signal sparsity, respectively. Experimental results upon ECG signals show that the proposed method, at a compression ratio of 10×, successfully reduces the isometry constant of the trained sensing matrices by 86% against random matrices and improves the overall reconstructed signal-to-noise ratio by 15dB over conventional model-driven approaches.

AB - Compressive sensing (CS) is a promising technology for realizing energy-efficient wireless sensors for long-term health monitoring. However, conventional model-driven CS frameworks suffer from limited compression ratio and reconstruction quality when dealing with physiological signals due to inaccurate models and the overlook of individual variability. In this paper, we propose a data-driven CS framework that can learn signal characteristics and personalized features from any individual recording of physiologic signals to enhance CS performance with a minimized number of measurements. Such improvements are accomplished by a co-training approach that optimizes the sensing matrix and the dictionary towards improved restricted isometry property and signal sparsity, respectively. Experimental results upon ECG signals show that the proposed method, at a compression ratio of 10×, successfully reduces the isometry constant of the trained sensing matrices by 86% against random matrices and improves the overall reconstructed signal-to-noise ratio by 15dB over conventional model-driven approaches.

KW - Data-driven compressive sensing

KW - Internet of things (IoT)

KW - mobile healthcare

KW - wearable sensing

UR - http://www.scopus.com/inward/record.url?scp=85023754121&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85023754121&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2017.7952278

DO - 10.1109/ICASSP.2017.7952278

M3 - Conference contribution

AN - SCOPUS:85023754121

SP - 861

EP - 865

BT - 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -