Fall Detection Threshold Adjustment via Walking Pattern Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic devices lack an effective method for detecting falls in users, particularly in varying walking environments, which can lead to inaccurate fall detection and delayed user intervention.
Innovation Solution
An electronic device equipped with sensors, memory, and a processor that obtains information on a user's walking pattern and environment, calculates differences between the user's walking pattern and reference/learned patterns, and sets a threshold for fall detection based on these differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional fixed-threshold fall detection is used, then the device complexity is low, but the measurement precision of fall detection deteriorates in varying walking environments
Solution Approach 1:
The patent implements dynamic threshold adjustment by continuously learning the user's walking patterns through sensors and adapting the fall detection threshold based on identified walking environments (indoor/outdoor, uphill/downhill). The processor dynamically modifies detection parameters according to real-time sensor data and environmental context, transforming a static detection system into an adaptive one that maintains high accuracy across varying conditions without requiring complex hardware changes
Solution Approach 2:
The system changes detection parameters (threshold values, sensitivity levels) based on identified walking environments and user-specific patterns. By analyzing sensor data to determine walking context and comparing against learned reference patterns, the system adjusts detection parameters to match current conditions, thereby improving measurement precision while keeping the underlying detection mechanism relatively simple
2Measurement precision
If user-specific walking patterns are learned and compared, then the fall detection accuracy improves, but the loss of time for pattern learning and processing increases
Solution Approach 1:
The system performs preliminary learning of user walking patterns during normal usage periods when the user is actively walking. By accumulating and analyzing sensor data during these periods, the system builds a reference profile of the user's gait characteristics in advance. This preliminary action allows the system to have detection-ready patterns stored, reducing the time needed for real-time analysis while maintaining high accuracy
Solution Approach 2:
The learning and detection processes operate continuously and concurrently rather than sequentially. The system continues to learn and refine walking patterns in the background while simultaneously performing fall detection. This continuous operation ensures that the useful action of pattern recognition is maintained without interruption, minimizing the time loss associated with periodic re-learning while sustaining high detection accuracy
Data Source
AI summary
According to an example, the electronic device includes at least one sensor, a memory, and a processor. The processor is configured to obtain first information related to a first walking pattern of a user of the electronic device and second information related to a walking environment of the user. The processor is configured to identify, based on the first information and third information related to a reference walking pattern which is stored in a memory, a first value indicating a difference between the first walking pattern and the reference walking pattern. The processor is configured to identify a second value indicating a difference between the first walking pattern and the second walking pattern. The processor is configured to identify a threshold value related to fall detection based on the first value, the second value, and the second information. The processor is configured to execute a function for the fall detection.