Accelerometer Step Counting via Most Active Axis Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional step counting devices struggle to accurately record gait status and count steps due to the need for precise orientation alignment with gravitational force, inability to differentiate between different gaits, and false measurements caused by motion noise, leading to incorrect step counts.
Innovation Solution
A wearable device equipped with an accelerometer that identifies the most active axis based on accelerometer data and applies a derivative operator to count peaks, using adaptive thresholds and peak elimination algorithms to accurately classify activity and count steps regardless of device orientation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional step counting devices require the user to position the device in a limited set of orientations aligned with gravitational force, then the device can function, but the ease of operation deteriorates because the user must maintain precise orientation alignment
Solution Approach 1:
The device dynamically identifies the most active axis from accelerometer data in real-time, allowing the device to adapt to any orientation worn by the user. This eliminates the need for fixed orientation alignment while maintaining measurement accuracy through adaptive signal processing.
Solution Approach 2:
The system changes the reference parameter from a fixed gravitational alignment to a dynamically determined most active axis based on motion patterns. This allows the device to function accurately regardless of how it is positioned on the user's body.
2Measurement precision
If existing devices use simple peak detection algorithms, then the device complexity is low, but the measurement precision deteriorates due to inability to differentiate signals corresponding to different gaits and false step measurements
Solution Approach 1:
The signal processing is divided into distinct stages: identifying the most active axis, applying derivative operators to enhance peaks, detecting peaks in the derivative signal, and eliminating duplicate peaks. This segmentation allows complex processing to be managed systematically while improving accuracy.
Solution Approach 2:
The derivative operator acts as an intermediary that transforms the raw accelerometer signal into a form where peaks are more distinct and easier to detect accurately. This intermediate transformation step enhances the detectability of step events while filtering out noise.
3Reliability
If existing devices count all detected peaks as steps, then the counting process is simple, but the reliability deteriorates due to overcounting caused by motion noise and inconsistent signal peaks
Solution Approach 1:
The system uses feedback by comparing the time difference between consecutive peaks to determine whether a detected peak represents a valid step or should be eliminated as a duplicate. This feedback mechanism improves reliability by correcting for overcounting.
Solution Approach 2:
The algorithm extracts and eliminates duplicate peaks by identifying peaks that occur too close together in time. This extraction of invalid peaks from the total count improves the reliability of step measurements.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution enables accurate classification of user activity and step counting by identifying the most active axis, smoothing data, and applying derivative operators to eliminate noise, resulting in improved accuracy and reliability of step counting.
Implementation Method 1
an accelerometer for measuring accelerometer data for a plurality of axes
Data Source
AI summary
Device and method for classifying the activity and/or counting steps of a user. A method for classifying the activity of a user can comprise measuring accelerometer data for a plurality of axes; identifying a most active one of the plurality of axes based on the accelerometer data; and classifying the activity of the user based on a signal amplitude of the accelerometer data for the most active axis and one or more threshold values.