Multi-device Activity State Recognition via Sensor Fusion

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing electronic devices struggle to accurately determine a user's activity state, leading to false or non-recognition of falls due to misidentification of activity states such as walking, biking, or vehicle boarding.

Innovation Solution

A system comprising a first electronic device and a second electronic device, where the first device requests the second device to identify a detected activity state, and updates the user's activity state based on the second device's identification, using sensors and communication modules to accurately determine walking, biking, or vehicle boarding states.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a single electronic device uses sensor data to determine user activity state, then the device can operate independently, but the activity state recognition accuracy deteriorates due to misidentification in complex scenarios

Engineering Contradiction:

Improveactivity state recognition accuracyVSAvoidsystem structure

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent combines multiple electronic devices (wearable device and terminal device) into a collaborative system for activity state recognition. The wearable device collects sensor data while the terminal device performs comprehensive analysis using multiple algorithms, merging their capabilities to achieve higher recognition accuracy than either device could achieve alone.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The terminal device acts as an intermediary that receives raw sensor data from the wearable device, processes it through multiple recognition algorithms, and determines the final activity state. This intermediary processing layer resolves the contradiction by offloading complex recognition tasks from the resource-constrained wearable device to the more powerful terminal device.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Device complexity

If the wearable device independently determines activity state using local sensors, then the system operates simply, but false fall detections occur due to misrecognition of activity states like biking or vehicle boarding

Engineering Contradiction:

Improvesystem structureVSAvoidfall detection accuracy

Core Design Contradiction:

Device complexityVSReliability

Solution Approach 1:

The system implements feedback by continuously comparing sensor data with multiple recognition algorithms and updating activity state determination based on this comparison. The terminal device receives sensor data from the wearable device, analyzes it through various algorithms, and provides corrected activity state information back to prevent false fall detections.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

Instead of relying on a single simple algorithm, the system employs multiple recognition algorithms (excessive action) including acceleration-based, gyro-based, and composite algorithms. This over-engineered approach ensures that even if one algorithm fails or misinterprets sensor data, other algorithms can compensate and provide accurate activity state recognition.

Inventive Principle:

Principle #16Partial or excessive action

3Measurement precision

If multiple recognition algorithms are deployed in the wearable device, then activity state recognition improves, but the device's computational load and power consumption increase

Engineering Contradiction:

Improveactivity state recognition accuracyVSAvoidpower consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent segments the computational workload by dividing it between the wearable device and the terminal device. The wearable device performs initial sensor data collection and basic processing, then transmits data to the terminal device for comprehensive multi-algorithm analysis. This segmentation allows multiple algorithms to run without overburdening the wearable device's limited computational resources.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The terminal device serves as an intermediary that handles the computationally intensive multi-algorithm processing. It receives sensor data from the wearable device, executes multiple recognition algorithms, and returns the final activity state determination. This intermediary architecture enables high-precision recognition while keeping the wearable device's power consumption low.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20250076337A1Method and electronic device for determining user activity state on basis of plurality of electronic devices

Publication Date: 2025.03.06 SAMSUNG ELECTRONICS CO LTD

AI summary

A system for determining an activity state of a user includes: a first electronic device; and a second electronic device, where the first electronic device is configured to: request the second electronic device to identify a first activity state detected as an activity state of a user in first electronic device, receive a second activity state corresponding to an identification result of identifying the first activity state in the second electronic device received from the second electronic device in response to the request and update the activity state of the user based on the second activity state, and where the second electronic device is configured to: based on receiving the request for identifying the first activity state from the first electronic device, identify whether the activity state of the user is the first activity state and transmit the second activity state corresponding to the identification result to the first electronic device.