Foot Step Counting via Acceleration Signal Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wearable sports devices cannot implement foot step counting when worn on the foot, as they are designed for wrist and waist use, lacking the necessary functionality for foot-based step detection.
Innovation Solution
A method and apparatus that collect and process acceleration information to count foot steps by combining data points into segments, extracting feature points, analyzing stable and fluctuation stages, and eliminating interference periods to determine the number of wave crests, which corresponds to the number of steps, using a portable device with a sensor, processor, and memory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing wearable sports devices are worn on the wrist or waist, then step counting can be implemented, but foot step counting cannot be implemented
Solution Approach 1:
The patent segments the step counting function from the wearing position constraint. By developing position-specific algorithms (wrist algorithm, waist algorithm, and foot algorithm), the system can be worn on different body parts while maintaining accurate step counting. The foot-specific algorithm processes acceleration data differently based on the unique motion characteristics of foot movement, enabling reliable foot step counting.
Solution Approach 2:
The patent applies local quality by creating position-specific processing characteristics. Each wearing position (wrist, waist, foot) has its own algorithm with parameters optimized for that location. The foot algorithm uses specific thresholds and processing methods tailored to foot acceleration patterns, ensuring accurate step detection regardless of where the device is worn.
2Adaptability or versatility
If foot step counting is implemented by processing acceleration signals, then foot-worn device functionality is enabled, but signal interference from foot motion characteristics increases
Solution Approach 1:
The patent extracts and eliminates interference periods from the acceleration signal. By identifying and removing segments of the signal that contain noise or irrelevant motion patterns, the algorithm isolates the genuine step-related acceleration data. This extraction of useful information from noisy data enables accurate foot step counting despite the complex motion characteristics of foot movement.
Solution Approach 2:
The patent converts the harmful effect of foot motion interference into a beneficial feature. By analyzing the characteristic patterns of foot acceleration during different phases of gait (stable stage and fluctuation stage), the algorithm uses the interference patterns themselves as indicators of step events. The previously harmful noise becomes a useful signal for detecting steps.
3Measurement precision
If combined acceleration processing is applied to foot-worn device data, then foot step counting accuracy reaches 94%, but processing complexity increases
Solution Approach 1:
The patent employs periodic action by dividing the continuous acceleration signal into discrete data segments with a specific number of points. This segmentation allows the algorithm to process the signal in manageable chunks, applying consistent processing steps to each segment. The periodic structure simplifies the complex processing by creating a repeatable pattern of analysis that can be efficiently implemented.
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
Enables accurate foot step counting with an accuracy of approximately 94% by processing acceleration signals from a foot-worn device, effectively addressing the limitation of existing technologies that cannot count foot steps.
Implementation Method 1
collecting an original signal in a specified time interval, and performing combined acceleration processing on the original signal
Data Source
AI summary
A method includes collecting an original signal in a specified time interval, performing combined acceleration processing on the original signal to obtain a combined acceleration signal, combining n data points of the combined acceleration signal into one data segment, extracting a feature point of the data segment, transforming feature points of all data segments into a feature point signal, analyzing the feature point signal to obtain a stable stage and a fluctuation stage of the feature point signal, performing interference period elimination on the feature point signal with the stable stage and the fluctuation stage according to a specified motion rule to obtain a second feature point signal, and counting a quantity of wave crests of the second feature point signal, where the quantity of wave crests is a quantity of steps.