Wearable Device EMG Signal Processing for Motion Artifact Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wearable devices for fitness tracking provide limited insight and often require manual input to measure exercise metrics, and they struggle with accurately analyzing electromyography (EMG) signals due to motion artifacts and signal cross-talk, which affects the accuracy of muscle activity monitoring.
Innovation Solution
A wearable device system that uses sensors to capture acceleration signals, filters and processes them to generate motion data, and employs Hidden Markov Modeling and feature-based classification to identify exercise activities, while also accounting for skin impedance and bio-impedance to improve signal quality and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If wearable devices use sensors to capture EMG signals for muscle activity monitoring, then the ability to track physical activities is improved, but motion artifacts and signal cross-talk reduce measurement precision
Solution Approach 1:
The patent introduces intermediary processing steps including signal filtering, motion artifact detection, and cross-talk correction algorithms. These intermediaries mediate between the raw sensor signals and the final muscle activity measurements, removing harmful components while preserving useful information.
Solution Approach 2:
The system dynamically adjusts signal processing parameters based on detected motion states and signal quality metrics. By changing filtering thresholds, integration windows, and correction factors in response to varying conditions, the system maintains measurement precision across different exercise intensities and motion patterns.
2Loss of information
If the system processes multiple sensor signals in real-time to provide detailed exercise metrics, then the insight provided is improved, but the device complexity increases
Solution Approach 1:
The patent divides the complex signal processing task into distinct segments: acceleration signal filtering, position computation, motion artifact identification, EMG signal processing, and integrated exercise metric calculation. Each segment handles a specific aspect of the data, making the overall system more manageable and implementable on wearable devices with limited processing power.
3Measurement precision
If manual input is required to measure exercise metrics, then the measurement capability is improved, but the ease of operation decreases
Solution Approach 1:
The system automatically performs exercise metric measurement without requiring manual user input. The sensors continuously capture data, and the processing algorithms automatically compute exercise metrics, eliminating the need for users to manually log or input exercise information while maintaining accurate measurement capability.
Data Source
AI summary
Systems, methods and devices for recognizing user activity using data from accelerometer or other sensors. “Feature-based” approach and “model-based” approaches are described. In a feature-based approach, various values are extracted from input signals and projected onto a space that is selected to facilitate better segregation of data points. Classifiers identify the regions in this projected space in which the data points fall to distinguish between the different activity types. In a “model-based” approach, a generative model is trained for each activity type. Different activity types can be distinguished by identifying similarities between the input data with the generative models.