Weighted Multi-Wearable Fall Detection for Standing and Sitting Users
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wearable devices struggle to accurately detect falls, particularly when a user is in a standing or sitting position, due to variations in impact and altitude changes, leading to inconsistent detection accuracy.
Innovation Solution
A method involving interworking between a walking assistance device and a watch-type wearable electronic device, utilizing a communication circuit, sensors, and processors to determine user posture and apply weighted analysis of sensor data from both devices to enhance fall detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single wearable device is used for fall detection, then the device complexity is reduced, but the fall detection accuracy deteriorates due to inability to account for different user postures and impact variations
Solution Approach 1:
The system divides the fall detection function across multiple wearable devices (e.g., walking assistance device on lower body and watch-type device on wrist). Each device independently detects fall-related parameters and transmits data to a server, which performs centralized analysis. This segmentation allows each device to remain relatively simple while achieving high overall detection accuracy through coordinated multi-device operation.
Solution Approach 2:
The system combines fall detection data from multiple wearable devices with different sensor configurations and detection capabilities. The server integrates information from both devices, applying weighted analysis based on device type and detected parameters. This merging of data sources compensates for individual device limitations and improves overall fall detection accuracy across various user postures.
2Measurement precision
If multiple electronic devices are used for fall detection, then the fall detection accuracy improves through complementary sensor data, but the device complexity and data processing requirements increase
Solution Approach 1:
A server acts as an intermediary between multiple wearable devices and the user. The server receives fall-related parameters from both the walking assistance device and the watch-type device, performs centralized weighted analysis, and generates fall detection results. This intermediary approach simplifies individual device complexity while enabling sophisticated multi-device data integration and analysis.
Solution Approach 2:
The system dynamically assigns weights to data from different devices based on their respective capabilities and the specific detection scenario. The walking assistance device may be assigned higher weight for detecting falls from standing position, while the watch-type device receives higher weight for detecting falls from sitting position. This dynamic weighting optimizes detection accuracy while managing system complexity.
3Measurement precision
If fall detection relies solely on inertial sensor impact measurements, then the device structure is simplified, but the detection accuracy deteriorates when users fall from sitting position or lose consciousness without wrist movement
Solution Approach 1:
The system employs multiple types of sensors across different devices to perform fall detection universally, regardless of user posture or consciousness state. The walking assistance device uses inertial sensors and atmospheric pressure sensors to detect falls from any position, while the watch-type device uses inertial sensors, atmospheric pressure sensors, and biometric sensors to detect falls and monitor consciousness. This multi-functional sensor configuration ensures comprehensive fall detection coverage.
Solution Approach 2:
The system changes the parameters being measured by incorporating atmospheric pressure data alongside inertial sensor data. Atmospheric pressure sensors detect altitude changes that occur during falls, providing an additional detection dimension that complements impact-based inertial measurements. This parameter expansion improves detection accuracy for falls from sitting position and unconscious falls without requiring complex additional hardware.
4Measurement precision
If fall detection uses atmospheric pressure sensor data, then the detection accuracy for altitude changes improves, but the energy consumption and device complexity increase
Solution Approach 1:
The system uses atmospheric pressure sensors partially, activating them only when fall detection is needed rather than continuously. The sensors are triggered based on inertial sensor detections or specific time intervals, reducing energy consumption while maintaining detection accuracy when falls occur. This partial action approach balances energy usage with detection precision.
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
Improves fall detection accuracy by leveraging the strengths of multiple devices based on user posture, ensuring precise identification of falls regardless of the user's position.
Implementation Method 1
a change in the user's altitude measured through an atmospheric pressure sensor
Implementation Method 2
fall impact (and a change in the waist angle) measured through an inertial sensor
Data Source
AI summary
An electronic device includes: a communication circuit; a first sensor; a processor; and memory storing instructions that cause the electronic device to: determine whether a user wearing the electronic device is in a standing posture; determine that the electronic device is a main electronic device and an external electronic device connected to the electronic device is a sub electronic device, based on determining that the user is in the standing posture; obtain first information related to a fall of the user; receive, from the external electronic device through the communication circuit, second information related to the fall of the user. The second information is obtained by the external electronic device; and detect the fall of the user, by applying a first weight to the first information and by applying a second weight corresponding to the sub electronic device. The second weight is lower than the first weight.