A 1.06-μW Smart ECG Processor in 65-nm CMOS for Real-Time Biometric Authentication and Personal Cardiac Monitoring

Shihui Yin, Minkyu Kim, Deepak Kadetotad, Yang Liu, Chisung Bae, Sang Joon Kim, Yu Cao, Jae-sun Seo

Research output: Contribution to journalArticle

Abstract

Many wearable devices employ the sensors for physiological signals (e.g., electrocardiogram or ECG) to continuously monitor personal health (e.g., cardiac monitoring). Considering private medical data storage, secure access to such wearable devices becomes a crucial necessity. Exploiting the ECG sensors present on wearable devices, we investigate the possibility of using ECG as the individually unique source for device authentication. In particular, we propose to use ECG features toward both cardiac monitoring and neural-network-based biometric authentication. For such complex functionalities to be seamlessly integrated in wearable devices, an accurate algorithm must be implemented with ultralow power and a small form factor. In this paper, a smart ECG processor is presented for ECG-based authentication as well as cardiac monitoring. Data-driven Lasso regression and low-precision techniques are developed to compress neural networks for feature extraction by 24.4×. The 65-nm testchip consumes 1.06 μW at 0.55 V for real-time ECG authentication. For authentication, equal error rates of 1.70%/2.18%/2.48% (best/average/worst) are achieved on the in-house 645-subject database. For cardiac monitoring, 93.13% arrhythmia detection sensitivity and 89.78% specificity are achieved for 42 subjects in the MIT-BIH arrhythmia database.

Original languageEnglish (US)
Article number8713394
Pages (from-to)2316-2326
Number of pages11
JournalIEEE Journal of Solid-State Circuits
Volume54
Issue number8
DOIs
StatePublished - Aug 1 2019

Fingerprint

Biometrics
Electrocardiography
Authentication
Monitoring
Neural networks
Sensors
Feature extraction
Health
Data storage equipment

Keywords

  • Arrhythmia detection
  • biometric authentication
  • cardiac monitoring
  • ECG
  • Lasso regression
  • sparse neural network (NN)
  • weight compression

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

A 1.06-μW Smart ECG Processor in 65-nm CMOS for Real-Time Biometric Authentication and Personal Cardiac Monitoring. / Yin, Shihui; Kim, Minkyu; Kadetotad, Deepak; Liu, Yang; Bae, Chisung; Kim, Sang Joon; Cao, Yu; Seo, Jae-sun.

In: IEEE Journal of Solid-State Circuits, Vol. 54, No. 8, 8713394, 01.08.2019, p. 2316-2326.

Research output: Contribution to journalArticle

Yin, Shihui ; Kim, Minkyu ; Kadetotad, Deepak ; Liu, Yang ; Bae, Chisung ; Kim, Sang Joon ; Cao, Yu ; Seo, Jae-sun. / A 1.06-μW Smart ECG Processor in 65-nm CMOS for Real-Time Biometric Authentication and Personal Cardiac Monitoring. In: IEEE Journal of Solid-State Circuits. 2019 ; Vol. 54, No. 8. pp. 2316-2326.
@article{abf5cf99c4e143b2951c81e10f073846,
title = "A 1.06-μW Smart ECG Processor in 65-nm CMOS for Real-Time Biometric Authentication and Personal Cardiac Monitoring",
abstract = "Many wearable devices employ the sensors for physiological signals (e.g., electrocardiogram or ECG) to continuously monitor personal health (e.g., cardiac monitoring). Considering private medical data storage, secure access to such wearable devices becomes a crucial necessity. Exploiting the ECG sensors present on wearable devices, we investigate the possibility of using ECG as the individually unique source for device authentication. In particular, we propose to use ECG features toward both cardiac monitoring and neural-network-based biometric authentication. For such complex functionalities to be seamlessly integrated in wearable devices, an accurate algorithm must be implemented with ultralow power and a small form factor. In this paper, a smart ECG processor is presented for ECG-based authentication as well as cardiac monitoring. Data-driven Lasso regression and low-precision techniques are developed to compress neural networks for feature extraction by 24.4×. The 65-nm testchip consumes 1.06 μW at 0.55 V for real-time ECG authentication. For authentication, equal error rates of 1.70{\%}/2.18{\%}/2.48{\%} (best/average/worst) are achieved on the in-house 645-subject database. For cardiac monitoring, 93.13{\%} arrhythmia detection sensitivity and 89.78{\%} specificity are achieved for 42 subjects in the MIT-BIH arrhythmia database.",
keywords = "Arrhythmia detection, biometric authentication, cardiac monitoring, ECG, Lasso regression, sparse neural network (NN), weight compression",
author = "Shihui Yin and Minkyu Kim and Deepak Kadetotad and Yang Liu and Chisung Bae and Kim, {Sang Joon} and Yu Cao and Jae-sun Seo",
year = "2019",
month = "8",
day = "1",
doi = "10.1109/JSSC.2019.2912304",
language = "English (US)",
volume = "54",
pages = "2316--2326",
journal = "IEEE Journal of Solid-State Circuits",
issn = "0018-9200",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "8",

}

TY - JOUR

T1 - A 1.06-μW Smart ECG Processor in 65-nm CMOS for Real-Time Biometric Authentication and Personal Cardiac Monitoring

AU - Yin, Shihui

AU - Kim, Minkyu

AU - Kadetotad, Deepak

AU - Liu, Yang

AU - Bae, Chisung

AU - Kim, Sang Joon

AU - Cao, Yu

AU - Seo, Jae-sun

PY - 2019/8/1

Y1 - 2019/8/1

N2 - Many wearable devices employ the sensors for physiological signals (e.g., electrocardiogram or ECG) to continuously monitor personal health (e.g., cardiac monitoring). Considering private medical data storage, secure access to such wearable devices becomes a crucial necessity. Exploiting the ECG sensors present on wearable devices, we investigate the possibility of using ECG as the individually unique source for device authentication. In particular, we propose to use ECG features toward both cardiac monitoring and neural-network-based biometric authentication. For such complex functionalities to be seamlessly integrated in wearable devices, an accurate algorithm must be implemented with ultralow power and a small form factor. In this paper, a smart ECG processor is presented for ECG-based authentication as well as cardiac monitoring. Data-driven Lasso regression and low-precision techniques are developed to compress neural networks for feature extraction by 24.4×. The 65-nm testchip consumes 1.06 μW at 0.55 V for real-time ECG authentication. For authentication, equal error rates of 1.70%/2.18%/2.48% (best/average/worst) are achieved on the in-house 645-subject database. For cardiac monitoring, 93.13% arrhythmia detection sensitivity and 89.78% specificity are achieved for 42 subjects in the MIT-BIH arrhythmia database.

AB - Many wearable devices employ the sensors for physiological signals (e.g., electrocardiogram or ECG) to continuously monitor personal health (e.g., cardiac monitoring). Considering private medical data storage, secure access to such wearable devices becomes a crucial necessity. Exploiting the ECG sensors present on wearable devices, we investigate the possibility of using ECG as the individually unique source for device authentication. In particular, we propose to use ECG features toward both cardiac monitoring and neural-network-based biometric authentication. For such complex functionalities to be seamlessly integrated in wearable devices, an accurate algorithm must be implemented with ultralow power and a small form factor. In this paper, a smart ECG processor is presented for ECG-based authentication as well as cardiac monitoring. Data-driven Lasso regression and low-precision techniques are developed to compress neural networks for feature extraction by 24.4×. The 65-nm testchip consumes 1.06 μW at 0.55 V for real-time ECG authentication. For authentication, equal error rates of 1.70%/2.18%/2.48% (best/average/worst) are achieved on the in-house 645-subject database. For cardiac monitoring, 93.13% arrhythmia detection sensitivity and 89.78% specificity are achieved for 42 subjects in the MIT-BIH arrhythmia database.

KW - Arrhythmia detection

KW - biometric authentication

KW - cardiac monitoring

KW - ECG

KW - Lasso regression

KW - sparse neural network (NN)

KW - weight compression

UR - http://www.scopus.com/inward/record.url?scp=85069831879&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069831879&partnerID=8YFLogxK

U2 - 10.1109/JSSC.2019.2912304

DO - 10.1109/JSSC.2019.2912304

M3 - Article

VL - 54

SP - 2316

EP - 2326

JO - IEEE Journal of Solid-State Circuits

JF - IEEE Journal of Solid-State Circuits

SN - 0018-9200

IS - 8

M1 - 8713394

ER -