Answers are grounded in this paper and cite their sources.
Wearable biomechanical signals exhibit substantial heterogeneity in frequency composition, temporal structure, and inter-individual variability, challenging low-complexity single-channel sensing front-ends across diverse tasks. We propose a multimodal biomechanical sensing framework integrating a low-complexity flexible triboelectric patch with task-adaptive computational decoding. The flexible patch exhibited a force sensitivity of 3.79 V N⁻¹ within the 1–5 N range, a device-to-device peak-voltage coefficient of variation of 3.11%, and stable output over 10,000 controlled loading cycles. Using the same front-end, task-specific pipelines were developed for three applications. For wrist-based physiological sensing in three volunteers, digital filtering separated respiration- and pulse-related components from composite signals, and the derived heart and respiratory rates showed parameter-level agreement with commercial reference devices, with R² values of 0.960 and 0.828, respectively. For throat-vibration voice recognition in three volunteers, full-record preprocessing, automatic utterance detection, multiscale feature extraction, and training-only model selection achieved 94.35% accuracy and a 94.33% macro-F1 score for 15 voice commands under subject-dependent bidirectional strict temporal holdout using data acquired under controlled laboratory conditions. For shoulder-motion recognition in five volunteers, amplitude-independent causal segmentation and multi-domain feature decoding were evaluated using a leave-one-subject-out protocol, and rapid personalization with four labeled cycles per motion class achieved 95.86% accuracy and a 95.79% macro-F1 score. These results support the feasibility of extending the multitask analytical capability of a low-complexity single-channel triboelectric sensing front-end through task-adaptive computational decoding without increasing hardware complexity, and suggest a promising hardware–software co-design strategy for multimodal wearable sensing.
No takes yet. Share an insight, caveat, or question.
Yang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: