Neuromorphic Classification of Geophone Signals with Legendre Memory Units

Avi Hazan; Shlomo Greenberg; Elishai Ezra Tsur

doi:10.1109/BioCAS67066.2025.00131

Neuromorphic Classification of Geophone Signals with Legendre Memory Units

Avi Hazan
, Shlomo Greenberg
, Elishai Ezra Tsur

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Geophones are acoustic detectors that respond to seismic waves, which generate ground vibrations. In this work, we utilized a neuromorphic (brain-inspired) Legendre Memory Units (LMUs)-driven neural model for low-power, real-time classification of Geophone data, discriminating vibrations, which were generated by human footsteps, a moving vehicle, and ambient noise. We show that our neuromorphic hardware-compatible neural design produces comparable results with state-of-the-art long-short-term memory (LSTM) models, achieving high test accuracy (LMU: 93.88%, LSTM: 95.15%). Our work highlights the potential of LMUs-driven inference models for classifying analog data in energy-constrained environments.

Original language	English
Title of host publication	Proceedings - 21st IEEE Biomedical Circuits and Systems, BioCAS 2025
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	552-556
Number of pages	5
ISBN (Electronic)	9798331573362
DOIs	https://doi.org/10.1109/BioCAS67066.2025.00131
State	Published - 2025
Event	21st IEEE Biomedical Circuits and Systems, BioCAS 2025 - Abu Dhabi, United Arab Emirates Duration: 16 Oct 2025 → 18 Oct 2025

Publication series

Name	Proceedings - 21st IEEE Biomedical Circuits and Systems, BioCAS 2025

Conference

Conference	21st IEEE Biomedical Circuits and Systems, BioCAS 2025
Country/Territory	United Arab Emirates
City	Abu Dhabi
Period	16/10/25 → 18/10/25

Bibliographical note

Publisher Copyright:
© 2025 IEEE.

Keywords

Edge inference
LMU
LSTM
Neuromorphic computing
Vibration classification

Access to Document

10.1109/BioCAS67066.2025.00131

Cite this

Hazan, A., Greenberg, S., & Tsur, E. E. (2025). Neuromorphic Classification of Geophone Signals with Legendre Memory Units. In Proceedings - 21st IEEE Biomedical Circuits and Systems, BioCAS 2025 (pp. 552-556). (Proceedings - 21st IEEE Biomedical Circuits and Systems, BioCAS 2025). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BioCAS67066.2025.00131

@inproceedings{21f007ac84534751a02aa5205ddc4296,

title = "Neuromorphic Classification of Geophone Signals with Legendre Memory Units",

abstract = "Geophones are acoustic detectors that respond to seismic waves, which generate ground vibrations. In this work, we utilized a neuromorphic (brain-inspired) Legendre Memory Units (LMUs)-driven neural model for low-power, real-time classification of Geophone data, discriminating vibrations, which were generated by human footsteps, a moving vehicle, and ambient noise. We show that our neuromorphic hardware-compatible neural design produces comparable results with state-of-the-art long-short-term memory (LSTM) models, achieving high test accuracy (LMU: 93.88\%, LSTM: 95.15\%). Our work highlights the potential of LMUs-driven inference models for classifying analog data in energy-constrained environments.",

keywords = "Edge inference, LMU, LSTM, Neuromorphic computing, Vibration classification",

author = "Avi Hazan and Shlomo Greenberg and Tsur, \{Elishai Ezra\}",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 21st IEEE Biomedical Circuits and Systems, BioCAS 2025 ; Conference date: 16-10-2025 Through 18-10-2025",

year = "2025",

doi = "10.1109/BioCAS67066.2025.00131",

language = "אנגלית",

series = "Proceedings - 21st IEEE Biomedical Circuits and Systems, BioCAS 2025",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "552--556",

booktitle = "Proceedings - 21st IEEE Biomedical Circuits and Systems, BioCAS 2025",

address = "ארצות הברית",

}

Hazan, A, Greenberg, S & Tsur, EE 2025, Neuromorphic Classification of Geophone Signals with Legendre Memory Units. in Proceedings - 21st IEEE Biomedical Circuits and Systems, BioCAS 2025. Proceedings - 21st IEEE Biomedical Circuits and Systems, BioCAS 2025, Institute of Electrical and Electronics Engineers Inc., pp. 552-556, 21st IEEE Biomedical Circuits and Systems, BioCAS 2025, Abu Dhabi, United Arab Emirates, 16/10/25. https://doi.org/10.1109/BioCAS67066.2025.00131

Neuromorphic Classification of Geophone Signals with Legendre Memory Units. / Hazan, Avi; Greenberg, Shlomo; Tsur, Elishai Ezra.
Proceedings - 21st IEEE Biomedical Circuits and Systems, BioCAS 2025. Institute of Electrical and Electronics Engineers Inc., 2025. p. 552-556 (Proceedings - 21st IEEE Biomedical Circuits and Systems, BioCAS 2025).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Neuromorphic Classification of Geophone Signals with Legendre Memory Units

AU - Hazan, Avi

AU - Greenberg, Shlomo

AU - Tsur, Elishai Ezra

PY - 2025

Y1 - 2025

N2 - Geophones are acoustic detectors that respond to seismic waves, which generate ground vibrations. In this work, we utilized a neuromorphic (brain-inspired) Legendre Memory Units (LMUs)-driven neural model for low-power, real-time classification of Geophone data, discriminating vibrations, which were generated by human footsteps, a moving vehicle, and ambient noise. We show that our neuromorphic hardware-compatible neural design produces comparable results with state-of-the-art long-short-term memory (LSTM) models, achieving high test accuracy (LMU: 93.88%, LSTM: 95.15%). Our work highlights the potential of LMUs-driven inference models for classifying analog data in energy-constrained environments.

AB - Geophones are acoustic detectors that respond to seismic waves, which generate ground vibrations. In this work, we utilized a neuromorphic (brain-inspired) Legendre Memory Units (LMUs)-driven neural model for low-power, real-time classification of Geophone data, discriminating vibrations, which were generated by human footsteps, a moving vehicle, and ambient noise. We show that our neuromorphic hardware-compatible neural design produces comparable results with state-of-the-art long-short-term memory (LSTM) models, achieving high test accuracy (LMU: 93.88%, LSTM: 95.15%). Our work highlights the potential of LMUs-driven inference models for classifying analog data in energy-constrained environments.

KW - Edge inference

KW - LMU

KW - LSTM

KW - Neuromorphic computing

KW - Vibration classification

UR - https://www.scopus.com/pages/publications/105033229741

U2 - 10.1109/BioCAS67066.2025.00131

DO - 10.1109/BioCAS67066.2025.00131

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:105033229741

T3 - Proceedings - 21st IEEE Biomedical Circuits and Systems, BioCAS 2025

SP - 552

EP - 556

BT - Proceedings - 21st IEEE Biomedical Circuits and Systems, BioCAS 2025

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 21st IEEE Biomedical Circuits and Systems, BioCAS 2025

Y2 - 16 October 2025 through 18 October 2025

ER -

Neuromorphic Classification of Geophone Signals with Legendre Memory Units

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this