Real-time activity recognition from seismic signature via multi-scale symbolic time series analysis (MSTSA)

Soumalya Sarkar, Thyagaraju Damarla, Asok Ray

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

4 Citations (Scopus)

Abstract

Reliability of unattended ground sensors (UGS) to detect and classify different activities (e.g., walking and digging) is often limited by high false alarm rates, possibly due to the lack of robustness of the underlying algorithms in different environmental conditions (e.g., soil types and moisture contents for seismic sensors), inability to model large variations in the signature of a single activity and limitations of on-board computation. In this regard, a fast and robust multi-scale symbolic time series analysis (MSTSA) framework has been formulated to detect and classify human activities from seismic signatures. The building block of the proposed framework is built upon the concept of applying the short-length symbolic time-series online classifier (SSTOC) via Dirichlet-Compound-Multinomial model (DCM) construction. The algorithm operates on symbol sequences that are generated from seismic time-series and intermediate event class time-series at different time scales. These building blocks, with different window sizes, are cascaded in multiple layers for event detection and activity classification. A variety of experiments have been conducted in the field, which include realistic scenarios of different types of walking/digging. The results of experiments show that an accuracy of more than 90% and a false alarm of around 5% can be achieved in real time for activity detection and recognition.

Original languageEnglish (US)
Title of host publicationACC 2015 - 2015 American Control Conference
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages5818-5823
Number of pages6
ISBN (Electronic)9781479986842
DOIs
StatePublished - Jul 28 2015
Event2015 American Control Conference, ACC 2015 - Chicago, United States
Duration: Jul 1 2015Jul 3 2015

Publication series

NameProceedings of the American Control Conference
Volume2015-July
ISSN (Print)0743-1619

Other

Other2015 American Control Conference, ACC 2015
CountryUnited States
CityChicago
Period7/1/157/3/15

Fingerprint

Time series analysis
Time series
Sensors
Classifiers
Moisture
Experiments
Soils

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Cite this

Sarkar, S., Damarla, T., & Ray, A. (2015). Real-time activity recognition from seismic signature via multi-scale symbolic time series analysis (MSTSA). In ACC 2015 - 2015 American Control Conference (pp. 5818-5823). [7172251] (Proceedings of the American Control Conference; Vol. 2015-July). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ACC.2015.7172251
Sarkar, Soumalya ; Damarla, Thyagaraju ; Ray, Asok. / Real-time activity recognition from seismic signature via multi-scale symbolic time series analysis (MSTSA). ACC 2015 - 2015 American Control Conference. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 5818-5823 (Proceedings of the American Control Conference).
@inproceedings{f9f284e9d3124a658339cf4a02c8982c,
title = "Real-time activity recognition from seismic signature via multi-scale symbolic time series analysis (MSTSA)",
abstract = "Reliability of unattended ground sensors (UGS) to detect and classify different activities (e.g., walking and digging) is often limited by high false alarm rates, possibly due to the lack of robustness of the underlying algorithms in different environmental conditions (e.g., soil types and moisture contents for seismic sensors), inability to model large variations in the signature of a single activity and limitations of on-board computation. In this regard, a fast and robust multi-scale symbolic time series analysis (MSTSA) framework has been formulated to detect and classify human activities from seismic signatures. The building block of the proposed framework is built upon the concept of applying the short-length symbolic time-series online classifier (SSTOC) via Dirichlet-Compound-Multinomial model (DCM) construction. The algorithm operates on symbol sequences that are generated from seismic time-series and intermediate event class time-series at different time scales. These building blocks, with different window sizes, are cascaded in multiple layers for event detection and activity classification. A variety of experiments have been conducted in the field, which include realistic scenarios of different types of walking/digging. The results of experiments show that an accuracy of more than 90{\%} and a false alarm of around 5{\%} can be achieved in real time for activity detection and recognition.",
author = "Soumalya Sarkar and Thyagaraju Damarla and Asok Ray",
year = "2015",
month = "7",
day = "28",
doi = "10.1109/ACC.2015.7172251",
language = "English (US)",
series = "Proceedings of the American Control Conference",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "5818--5823",
booktitle = "ACC 2015 - 2015 American Control Conference",
address = "United States",

}

Sarkar, S, Damarla, T & Ray, A 2015, Real-time activity recognition from seismic signature via multi-scale symbolic time series analysis (MSTSA). in ACC 2015 - 2015 American Control Conference., 7172251, Proceedings of the American Control Conference, vol. 2015-July, Institute of Electrical and Electronics Engineers Inc., pp. 5818-5823, 2015 American Control Conference, ACC 2015, Chicago, United States, 7/1/15. https://doi.org/10.1109/ACC.2015.7172251

Real-time activity recognition from seismic signature via multi-scale symbolic time series analysis (MSTSA). / Sarkar, Soumalya; Damarla, Thyagaraju; Ray, Asok.

ACC 2015 - 2015 American Control Conference. Institute of Electrical and Electronics Engineers Inc., 2015. p. 5818-5823 7172251 (Proceedings of the American Control Conference; Vol. 2015-July).

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

TY - GEN

T1 - Real-time activity recognition from seismic signature via multi-scale symbolic time series analysis (MSTSA)

AU - Sarkar, Soumalya

AU - Damarla, Thyagaraju

AU - Ray, Asok

PY - 2015/7/28

Y1 - 2015/7/28

N2 - Reliability of unattended ground sensors (UGS) to detect and classify different activities (e.g., walking and digging) is often limited by high false alarm rates, possibly due to the lack of robustness of the underlying algorithms in different environmental conditions (e.g., soil types and moisture contents for seismic sensors), inability to model large variations in the signature of a single activity and limitations of on-board computation. In this regard, a fast and robust multi-scale symbolic time series analysis (MSTSA) framework has been formulated to detect and classify human activities from seismic signatures. The building block of the proposed framework is built upon the concept of applying the short-length symbolic time-series online classifier (SSTOC) via Dirichlet-Compound-Multinomial model (DCM) construction. The algorithm operates on symbol sequences that are generated from seismic time-series and intermediate event class time-series at different time scales. These building blocks, with different window sizes, are cascaded in multiple layers for event detection and activity classification. A variety of experiments have been conducted in the field, which include realistic scenarios of different types of walking/digging. The results of experiments show that an accuracy of more than 90% and a false alarm of around 5% can be achieved in real time for activity detection and recognition.

AB - Reliability of unattended ground sensors (UGS) to detect and classify different activities (e.g., walking and digging) is often limited by high false alarm rates, possibly due to the lack of robustness of the underlying algorithms in different environmental conditions (e.g., soil types and moisture contents for seismic sensors), inability to model large variations in the signature of a single activity and limitations of on-board computation. In this regard, a fast and robust multi-scale symbolic time series analysis (MSTSA) framework has been formulated to detect and classify human activities from seismic signatures. The building block of the proposed framework is built upon the concept of applying the short-length symbolic time-series online classifier (SSTOC) via Dirichlet-Compound-Multinomial model (DCM) construction. The algorithm operates on symbol sequences that are generated from seismic time-series and intermediate event class time-series at different time scales. These building blocks, with different window sizes, are cascaded in multiple layers for event detection and activity classification. A variety of experiments have been conducted in the field, which include realistic scenarios of different types of walking/digging. The results of experiments show that an accuracy of more than 90% and a false alarm of around 5% can be achieved in real time for activity detection and recognition.

UR - http://www.scopus.com/inward/record.url?scp=84940915411&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84940915411&partnerID=8YFLogxK

U2 - 10.1109/ACC.2015.7172251

DO - 10.1109/ACC.2015.7172251

M3 - Conference contribution

AN - SCOPUS:84940915411

T3 - Proceedings of the American Control Conference

SP - 5818

EP - 5823

BT - ACC 2015 - 2015 American Control Conference

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Sarkar S, Damarla T, Ray A. Real-time activity recognition from seismic signature via multi-scale symbolic time series analysis (MSTSA). In ACC 2015 - 2015 American Control Conference. Institute of Electrical and Electronics Engineers Inc. 2015. p. 5818-5823. 7172251. (Proceedings of the American Control Conference). https://doi.org/10.1109/ACC.2015.7172251