BARTON: Low Power Tongue Movement Sensing with In-ear Barometers

Balz Maag, Zimu Zhou, Olga Saukh, Lothar Thiele

Research output: Contribution to conferencePaperResearchpeer-review

Abstract

Sensing tongue movements enables various applications in hands-free interaction and alternative communication. We propose BARTON, a BARometer based low-power and robust TONgue movement sensing system. Using a low sampling rate of below 50 Hz, and only extracting simple temporal features from in-ear pressure signals, we demonstrate that it is plausible to distinguish important tongue gestures (left, right, forward) at low power consumption. We prototype BARTON with commodity earpieces integrated with COTS barometers for in-ear pressure sensing and an ARM micro-controller for signal processing. Evaluations show that BARTON yields 94% classification accuracy and 8.4mW power consumption, which achieves comparable accuracy, but consumes 44 times lower energy than the state-of-the-art microphone-based solutions. BARTON is also robust to head movements and operates with music played directly from earphones.
Original languageEnglish
Pages9-16
Number of pages8
DOIs
Publication statusPublished - 15 Dec 2017
Event23rd IEEE International Conference on Parallel and Distributed Systems (ICPADS) - Shenzen, China
Duration: 15 Dec 201717 Dec 2017
http://futurenet.szu.edu.cn/icpads2017/?index.html

Conference

Conference23rd IEEE International Conference on Parallel and Distributed Systems (ICPADS)
CountryChina
CityShenzen
Period15/12/1717/12/17
Internet address

Fingerprint

Barometers
Electric power utilization
Earphones
Microphones
Signal processing
Sampling
Controllers
Communication

Keywords

  • Human computer interaction
  • Ubiquitous computing
  • Pressure sensors

Cite this

Maag, B., Zhou, Z., Saukh, O., & Thiele, L. (2017). BARTON: Low Power Tongue Movement Sensing with In-ear Barometers. 9-16. Paper presented at 23rd IEEE International Conference on Parallel and Distributed Systems (ICPADS), Shenzen, China. https://doi.org/10.1109/ICPADS.2017.00013

BARTON: Low Power Tongue Movement Sensing with In-ear Barometers. / Maag, Balz; Zhou, Zimu; Saukh, Olga; Thiele, Lothar.

2017. 9-16 Paper presented at 23rd IEEE International Conference on Parallel and Distributed Systems (ICPADS), Shenzen, China.

Research output: Contribution to conferencePaperResearchpeer-review

Maag, B, Zhou, Z, Saukh, O & Thiele, L 2017, 'BARTON: Low Power Tongue Movement Sensing with In-ear Barometers' Paper presented at 23rd IEEE International Conference on Parallel and Distributed Systems (ICPADS), Shenzen, China, 15/12/17 - 17/12/17, pp. 9-16. https://doi.org/10.1109/ICPADS.2017.00013
Maag B, Zhou Z, Saukh O, Thiele L. BARTON: Low Power Tongue Movement Sensing with In-ear Barometers. 2017. Paper presented at 23rd IEEE International Conference on Parallel and Distributed Systems (ICPADS), Shenzen, China. https://doi.org/10.1109/ICPADS.2017.00013
Maag, Balz ; Zhou, Zimu ; Saukh, Olga ; Thiele, Lothar. / BARTON: Low Power Tongue Movement Sensing with In-ear Barometers. Paper presented at 23rd IEEE International Conference on Parallel and Distributed Systems (ICPADS), Shenzen, China.8 p.
@conference{3aab1b91b902472da4cf2fe36ca8fc2e,
title = "BARTON: Low Power Tongue Movement Sensing with In-ear Barometers",
abstract = "Sensing tongue movements enables various applications in hands-free interaction and alternative communication. We propose BARTON, a BARometer based low-power and robust TONgue movement sensing system. Using a low sampling rate of below 50 Hz, and only extracting simple temporal features from in-ear pressure signals, we demonstrate that it is plausible to distinguish important tongue gestures (left, right, forward) at low power consumption. We prototype BARTON with commodity earpieces integrated with COTS barometers for in-ear pressure sensing and an ARM micro-controller for signal processing. Evaluations show that BARTON yields 94{\%} classification accuracy and 8.4mW power consumption, which achieves comparable accuracy, but consumes 44 times lower energy than the state-of-the-art microphone-based solutions. BARTON is also robust to head movements and operates with music played directly from earphones.",
keywords = "Human computer interaction, Ubiquitous computing, Pressure sensors",
author = "Balz Maag and Zimu Zhou and Olga Saukh and Lothar Thiele",
year = "2017",
month = "12",
day = "15",
doi = "10.1109/ICPADS.2017.00013",
language = "English",
pages = "9--16",
note = "23rd IEEE International Conference on Parallel and Distributed Systems (ICPADS) ; Conference date: 15-12-2017 Through 17-12-2017",
url = "http://futurenet.szu.edu.cn/icpads2017/?index.html",

}

TY - CONF

T1 - BARTON: Low Power Tongue Movement Sensing with In-ear Barometers

AU - Maag, Balz

AU - Zhou, Zimu

AU - Saukh, Olga

AU - Thiele, Lothar

PY - 2017/12/15

Y1 - 2017/12/15

N2 - Sensing tongue movements enables various applications in hands-free interaction and alternative communication. We propose BARTON, a BARometer based low-power and robust TONgue movement sensing system. Using a low sampling rate of below 50 Hz, and only extracting simple temporal features from in-ear pressure signals, we demonstrate that it is plausible to distinguish important tongue gestures (left, right, forward) at low power consumption. We prototype BARTON with commodity earpieces integrated with COTS barometers for in-ear pressure sensing and an ARM micro-controller for signal processing. Evaluations show that BARTON yields 94% classification accuracy and 8.4mW power consumption, which achieves comparable accuracy, but consumes 44 times lower energy than the state-of-the-art microphone-based solutions. BARTON is also robust to head movements and operates with music played directly from earphones.

AB - Sensing tongue movements enables various applications in hands-free interaction and alternative communication. We propose BARTON, a BARometer based low-power and robust TONgue movement sensing system. Using a low sampling rate of below 50 Hz, and only extracting simple temporal features from in-ear pressure signals, we demonstrate that it is plausible to distinguish important tongue gestures (left, right, forward) at low power consumption. We prototype BARTON with commodity earpieces integrated with COTS barometers for in-ear pressure sensing and an ARM micro-controller for signal processing. Evaluations show that BARTON yields 94% classification accuracy and 8.4mW power consumption, which achieves comparable accuracy, but consumes 44 times lower energy than the state-of-the-art microphone-based solutions. BARTON is also robust to head movements and operates with music played directly from earphones.

KW - Human computer interaction

KW - Ubiquitous computing

KW - Pressure sensors

U2 - 10.1109/ICPADS.2017.00013

DO - 10.1109/ICPADS.2017.00013

M3 - Paper

SP - 9

EP - 16

ER -