Assessing the Resilience of sEMG Classifiers to Sensor Malfunction and Signal Saturation
Sensors · 13.04.2026 · 26 (8) · 2386
DOI: 10.3390/s26082386 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 13.04.2026. Verfasst von Congyi Zhang und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Computing, Mathematics and Physics, University of Portsmouth, Portsmouth PO1 3HE, UK. Hauptthemen laut MEDLINE: Electromyography, Signal Processing, Computer-Assisted. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Hangzhou Key Scientific Research Program.
Abstract (Originalsprache)
Surface electromyography (sEMG) is widely used for gesture recognition, yet the way classic feature–classifier pipelines fail under realistic signal degradations is still poorly quantified. Existing studies typically report accuracy on clean laboratory data, leaving open how amplitude saturation and channel dropout jointly affect different feature combinations, classifiers, and subjects. In this work, we provide, to our knowledge, the first systematic robustness map of a conventional sEMG pipeline under controlledclipping and single-sensor failure. sEMG from nine subjects performing a multi-session, multi-gesture protocol is windowed (250 ms, 50 ms hop) and represented using four common time-domain features (Root Mean Square, Variance, Zero Crossing, and Waveform Length). We exhaustively evaluated single features and all pairwise fusions with three standard classifiers (Support Vector Machine (RBF kernel), Linear Discriminant Analysis, and Random Forest) over (i) a sweep of symmetric saturation thresholds (10−6–10−1) and (ii) five single-channel dropout scenarios, reporting subject-wise dispersion rather than aggregate scores alone. This design enables explicit characterization of the following: (1) accuracy recovery as clipping weakens for each feature pair; (2) dependency of robustness on which channel fails; and (3) differences among Support Vector Machine, Linear Discriminant Analysis, and Random Forest under identical degradations. The results show that lightweight feature pairs (Root Mean Square + Waveform Length, Variance + Zero Crossing, and Waveform Length + Zero Crossing) coupled with Random Forest form a consistently robust operating point, with performance recovering as clipping weakens and remaining resilient under single-channel dropout. Beyond robustness, the conventional pipeline trains substantially faster than representative deep learning baselines under a unified end-to-end timing definition, supporting real-time recalibration and repeated robustness sweeps in wearable deployments.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Computing, Mathematics and Physics, University of Portsmouth, Portsmouth PO1 3HE, UK.
- School of Communication Engineering, Hangzhou Dianzi University, Hangzhou 310005, China.
ORCID-Profile: Congyi Zhang, Dalin Zhou, Yinfeng Fang
Förderung
- Hangzhou Key Scientific Research Program
Zitieren
APA
Zhang, C., Zhou, D., Fang, Y., Gao, D., & Ju, Z. (2026). Assessing the Resilience of sEMG Classifiers to Sensor Malfunction and Signal Saturation. Sensors, 26(8), 2386. https://doi.org/10.3390/s26082386
Vancouver
Zhang C, Zhou D, Fang Y, Gao D, Ju Z. Assessing the Resilience of sEMG Classifiers to Sensor Malfunction and Signal Saturation. Sensors. 2026;26(8):2386. doi:10.3390/s26082386
BibTeX
@article{Zhang2026,
author = {Zhang, Congyi and Zhou, Dalin and Fang, Yinfeng and Gao, Dongxu and Ju, Zhaojie},
title = {Assessing the Resilience of sEMG Classifiers to Sensor Malfunction and Signal Saturation},
journal = {Sensors},
year = {2026},
volume = {26},
number = {8},
pages = {2386},
doi = {10.3390/s26082386}
}
EUVIMED