Distributed Sensor Data Processing for Wearable Devices

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional wearable devices face challenges with high energy consumption, heat generation, and short battery life due to the need for energy-intensive processing of sensor data such as audio and image processing, which limits their usage duration.

Innovation Solution

A distributed processing system where a low-power wearable device offloads energy-intensive operations to a connected computing device or server via wireless connections, using smaller machine learning models for initial detection and transmitting data for further processing, allowing more complex analyses to be performed remotely.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If energy-intensive processing operations are performed locally in wearable devices, then processing capabilities and machine learning precision are improved, but power consumption increases and battery life decreases

Engineering Contradiction:

Improvemachine learning precisionVSAvoidpower consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The processing system is segmented into multiple devices: a wearable device that performs initial sound detection using a first ML model, and a computing device that performs further sound classification using a second ML model. This segmentation allows the wearable device to use minimal processing power while maintaining overall system precision through distributed intelligence.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

A computing device acts as an intermediary between the wearable device and the final processing outcome. The wearable device detects sounds of interest and transmits audio data to the computing device, which then performs the more energy-intensive further classification, thereby mediating the processing load away from the wearable device.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Power

If more circuit components including batteries are added to wearable devices, then processing capabilities are improved, but device weight increases causing discomfort

Engineering Contradiction:

Improveprocessing capabilityVSAvoiddevice weight

Core Design Contradiction:

PowerVSWeight of moving object

Solution Approach 1:

The heavy processing circuitry and large battery are extracted from the wearable device and placed in a separate computing device. The wearable device retains only essential components for capturing sensor data and running lightweight ML models, significantly reducing its weight while maintaining processing capability through the connected computing device.

Inventive Principle:

Principle #2Taking out (Extraction)

3Productivity

If energy-intensive operations are performed in wearable devices, then processing performance is improved, but heat generation increases making the device uncomfortable for extended wear

Engineering Contradiction:

Improveprocessing performanceVSAvoidheat generation

Core Design Contradiction:

ProductivityVSTemperature

Solution Approach 1:

The computing device serves as a thermal intermediary, absorbing the heat-generating processing operations. The wearable device performs only lightweight detection tasks that generate minimal heat, while the computing device handles the energy-intensive further classification in the background, effectively transferring the thermal burden away from the wearable device.

Inventive Principle:

Principle #24Intermediary (Mediator)

4Volume of moving object

If limited battery capacity is used in wearable devices, then device compactness is maintained, but usage duration becomes relatively short

Engineering Contradiction:

Improvedevice compactnessVSAvoidusage duration

Core Design Contradiction:

Volume of moving objectVSDuration of action of moving object

Solution Approach 1:

The system is segmented such that the wearable device maintains its compact form factor with limited battery capacity, while the computing device provides extended processing resources. The wearable device can operate for extended periods by performing only lightweight detection and offloading to the computing device when needed, effectively extending the system's operational duration without increasing the wearable device's size.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS12057126B2Distributed sensor data processing using multiple classifiers on multiple devices

Publication Date: 2024.08.06 GOOGLE LLC

AI summary

According to an aspect, a method for distributed sound/image recognition using a wearable device includes receiving, via at least one sensor device, sensor data, and detecting, by a classifier of the wearable device, whether or not the sensor data includes an object of interest. The classifier configured to execute a first machine learning (ML) model. The method includes transmitting, via a wireless connection, the sensor data to a computing device in response to the object of interest being detected within the sensor data, where the sensor data is configured to be used by a second ML model on the computing device or a server computer for further sound/image classification.