Probabilistic Cadence Estimation from Floor Vibration Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating cadence and walking speed in uncontrolled environments, such as home settings, face challenges due to variability in gait patterns, measurement biases, and inaccuracies in event identification, particularly with floor vibrations, leading to unreliable health assessments.
Innovation Solution
A probabilistic estimation method using sensor data sets to calculate cadence and walking speed, incorporating uncertainty calculations through Gaussian likelihood, and optional alert triggers for threshold exceedance, implemented in a system with preprocessing and estimation components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If floor vibrations are used to identify walking patterns in home environments, then non-intrusive monitoring is achieved, but measurement precision deteriorates due to signal-to-noise ratio issues and event identification inaccuracies
Solution Approach 1:
The patent introduces an intermediary probabilistic framework that mediates between the noisy floor vibration signals and the final gait parameter estimates. This framework includes uncertainty quantification and data fusion mechanisms that reconcile the conflicting requirements of non-intrusive monitoring and measurement precision.
Solution Approach 2:
The patent transforms the raw vibration signals into multiple derived parameters (cadence, speed, step length) with associated uncertainty estimates. By changing from direct event identification to probabilistic parameter estimation, the system maintains ease of operation while improving measurement precision through statistical methods.
2Reliability
If probabilistic estimation with uncertainty calculation is implemented, then reliability of health assessments improves, but device complexity increases
Solution Approach 1:
The patent segments the complex probabilistic estimation problem into distinct computational modules: signal preprocessing, event detection with uncertainty, parameter estimation, and final health assessment. This segmentation manages device complexity by organizing the computational burden into manageable, independent components.
Solution Approach 2:
The patent replaces complex mechanical sensor systems with computational complexity. Instead of using multiple sophisticated sensors to directly measure gait parameters, the system uses simpler floor vibration sensors combined with probabilistic algorithms, trading hardware simplicity for software complexity.
3Measurement precision
If multiple sensor data sets are collected and processed, then measurement precision improves through data fusion, but loss of time increases due to processing requirements
Solution Approach 1:
The patent performs preliminary processing of sensor data sets during data collection, pre-computing features and statistics that will be needed for final parameter estimation. This preliminary action reduces the computational burden during the actual estimation phase, minimizing time loss while maintaining measurement precision.
Solution Approach 2:
The patent implements continuous processing of sensor data streams, where each new data set is integrated with previous data without interruption. This continuous action maintains measurement precision through accumulating evidence while minimizing idle processing time between 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
Enhances the accuracy of cadence and walking speed estimation by accounting for uncertainty, reducing the impact of measurement variability and environmental factors, thereby improving the reliability of health assessments in unattended scenarios.
Implementation Method 1
Floor vibrations caused by the impact of footsteps during walking create deformations in the floor that sensors like accelerometers or geophones can detect
Data Source
AI summary
The system uses a probabilistic approach to extract cadence and walking speed from sensor data. This approach prevents the possibility of missing step events in the acceleration signal due to direct source collection or signal enhancement.