Exercise Assistance System Repetition Labeling via Signal Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional exercise assistance systems struggle to accurately detect repetitions in sensor signals from user movements due to noise interference, which limits their ability to provide personalized guidance and feedback.
Innovation Solution
An exercise assistance system that generates a combination signal from sensor data, applies a low-pass filter to smooth the signal, and uses label triplets to identify repetitions, with a classification model to differentiate between repetitions and noise, ensuring accurate labeling and feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If local extrema are used to detect repetitions in sensor signals, then the detection method is simple, but the accuracy deteriorates due to noise interference and missing extrema in noncontinuous repetitions
Solution Approach 1:
The patent introduces an intermediary signal processing pipeline between the raw sensor data and the repetition detection. This pipeline includes generating a combination signal from multiple sensor signals, applying a low-pass filter to reduce noise, and then detecting repetitions in this processed signal. The intermediary processing steps enable accurate repetition detection even when the raw signal contains noise or missing extrema.
2Reliability
If noise is present in sensor signals, then the signal represents real-world conditions, but the repetitions become obscured and difficult to detect
Solution Approach 1:
The patent extracts the repetition information from the noisy sensor signal through a systematic process. It combines multiple sensor signals to create a composite signal that emphasizes repetitive patterns, applies a low-pass filter to separate low-frequency repetition components from high-frequency noise, and then identifies repetitions in this cleaned signal. This extraction process isolates the useful repetition information from the harmful noise.
Solution Approach 2:
The patent changes the frequency parameter of the signal by applying a low-pass filter with a specific cutoff frequency. This parameter change allows the system to retain low-frequency repetition components while removing high-frequency noise. The cutoff frequency is carefully selected to preserve the repetition frequencies of interest while filtering out unwanted noise.
3Measurement precision
If multiple sensor signals are combined to accentuate repetitions, then the signal clarity improves, but the processing complexity increases
Solution Approach 1:
The patent merges multiple sensor signals into a single combination signal that accentuates repetition patterns. It combines the signals through summation and quadrature summation, then merges the results by taking the maximum or sum of the combined signals. This merging process consolidates information from multiple sensors to create a clearer representation of repetitions, making them easier to detect and label.
Data Source
AI summary
An exercise assistance system labels sensor signals received from a movement measurement device to identify repetitions within the sensor signals. The movement measurement device can transmit motion data describing the user's movement to the exercise assistance system, and includes the sensor signals generated by one or more sensors in the movement measurement device. The exercise assistance system generates a combination signal from the received sensor signals and passes the combination signal through a low pass filter to generate a labeling signal. The exercise assistance system can generate three types of labels for the labeling signal identifying different parts of repetitions. The exercise assistance system labels the sensor signals using the generated labels and can identify repetitions in the labeled sensor signals using a classification model.