Swimming Activity Monitoring Using Hidden Markov Model Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for monitoring swimming activity only provide quantitative data and fail to differentiate between various swimming types, limiting their usefulness for tracking progress and training.
Innovation Solution
A system using a waterproof housing with a motion sensor and hidden Markov model analysis to determine swimming types over time, employing a triaxial accelerometer or gyrometer, and featuring low-pass filtering to reduce noise, allowing differentiation between breaststroke, crawl, butterfly, and backstroke.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a motion sensor is used to monitor swimming activity, then quantitative monitoring is achieved, but qualitative analysis of swimming types cannot be performed
Solution Approach 1:
The patent segments the continuous swimming activity signal into discrete swimming stroke types (crawl, breaststroke, butterfly, backstroke) by analyzing acceleration patterns along multiple axes. The motion sensor data is divided into sequential segments that can be classified into distinct stroke categories, enabling both quantitative tracking and qualitative identification of swimming types.
Solution Approach 2:
The patent transitions from simple quantitative measurement to qualitative analysis by introducing dimensional analysis of acceleration vectors. By examining the direction and magnitude of acceleration along multiple axes (x, y, z), the system captures the distinctive patterns of different swimming strokes, adding a qualitative dimension to the monitoring data.
2Measurement precision
If complex analysis methods are used to differentiate swimming types, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent replaces complex mechanical or visual analysis systems with a streamlined electronic solution. A single motion sensor combined with algorithmic processing of acceleration data substitutes for more complex systems such as multiple sensors, video analysis equipment, or mechanical measurement devices, achieving accurate stroke identification with minimal hardware complexity.
Solution Approach 2:
The patent achieves swimming type differentiation by analyzing changes in acceleration parameters (magnitude, direction, frequency) rather than requiring complex system architecture. By monitoring how acceleration parameters vary throughout the swimming cycle, the system can identify distinctive stroke patterns using simple computational logic.
3Measurement precision
If multiple measurement axes are used to improve swimming type differentiation, then analysis accuracy improves, but data processing complexity increases
Solution Approach 1:
The patent performs preliminary organization of multi-axis acceleration data into meaningful patterns before full analysis. By pre-processing the raw sensor data to identify characteristic acceleration vectors and temporal patterns, the system reduces the complexity of subsequent stroke classification, making the processing more efficient and manageable.
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 qualitative analysis of swimming types, allowing users to track progress and compile statistics, improving the accuracy of swimming stroke identification.
Implementation Method 1
The system comprises a waterproof housing (1) comprising a motion sensor, in this instance a triaxial accelerometer (AT)
Implementation Method 2
The system comprises, furthermore, a low-pass filter of cutoff frequency lying between 0.5 Hz and 5 Hz
Data Source
AI summary
A system for observing a swimming activity of a person includes a waterproof housing (BET) having a motion sensor (MS), and is furnished with fixing means (BEL) for securely fastening the housing (BET) to a part of the body of a user. The system has analysis means (AN) for analyzing the signals transmitted by the motion sensor (MS) to at least one measurement axis and which are adapted for determining the type of swimming of the user as a function of time by using a hidden Markov model with N states corresponding respectively to N types of swimming.