Edge Sensor Data Compression Using Template Shape Matching

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Distributed computing systems face challenges in accurately and efficiently analyzing sensor data at resource-constrained edge nodes due to limited computational resources and bandwidth, leading to data degradation during compression and transfer.

Innovation Solution

The method involves clustering and shape matching techniques, where unique patterns in sensor data are identified and associated with specific events, allowing for probabilistic classification and selection of appropriate compression schemes, reducing data transmission to only the error between raw signals and known templates, utilizing dynamic time warping and k-nearest neighbor algorithms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If data is increasingly compressed at edge nodes to reduce bandwidth usage and storage requirements, then data transmission efficiency and resource utilization improve, but data accuracy and quality degrade after decompression

Engineering Contradiction:

Improvebandwidth usageVSAvoiddata accuracy

Core Design Contradiction:

Loss of energyVSMeasurement precision

Solution Approach 1:

The system performs preliminary clustering and template creation during a training phase before actual data compression. By pre-identifying event patterns and creating representative templates, the system enables efficient compression without sacrificing accuracy, as the templates capture the essential characteristics of sensor events

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Instead of compressing raw sensor data directly, the system creates simplified representations by matching data against pre-established templates. The compressed data includes only the template identifier and residual error information, effectively copying only the essential pattern characteristics rather than the full raw signal

Inventive Principle:

Principle #26Copying

2Productivity

If computational resources at edge nodes are increased to improve data processing capability, then data analysis accuracy improves, but device complexity and cost increase

Engineering Contradiction:

Improvedata processing capabilityVSAvoidcomputational resource requirements

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system segments the computational workload between edge nodes and centralized servers. Edge nodes perform only lightweight clustering and template matching, while more complex operations are handled centrally. This segmentation allows edge nodes to remain simple while still providing intelligent data processing capabilities

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system changes the parameter of computational complexity by using template-based matching instead of full data analysis at the edge. This approach reduces the computational parameters required at edge nodes while maintaining the ability to process and classify sensor data effectively

Inventive Principle:

Principle #35Parameter changes

3Productivity

If data transmission volume is reduced to optimize bandwidth utilization, then network efficiency improves, but information completeness and analysis quality deteriorate

Engineering Contradiction:

Improvenetwork efficiencyVSAvoidinformation completeness

Core Design Contradiction:

ProductivityVSLoss of information

Solution Approach 1:

The system extracts only the essential information from sensor data for transmission. By matching data against templates and transmitting only the template identifier and residual error, the system extracts the most critical information while discarding redundant details, thus optimizing bandwidth without losing essential event characteristics

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The template acts as an intermediary between raw sensor data and the compressed representation. Instead of directly compressing raw data, the system uses templates as an intermediate layer that captures event patterns, enabling efficient transmission while preserving the essential information needed for accurate reconstruction and analysis

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11431805B2Systems and methods for compressing sensor data using clustering and shape matching in edge nodes of distributed computing networks

Publication Date: 2022.08.30 SIGNIFY HOLDING BV

AI summary

A system and method for compressing sensor data at an edge node of a distributed computing network. The method includes training the edge node to with a plurality of known signal templates. Each known signal template corresponding to a corresponding one of a plurality of events observable by the sensor. A raw data signal is collected by a sensor of the edge node. The raw data signal is classified to one of the known signal templates based on a degree of similarity between the raw data signal and the known signal template. A compression scheme is selected based on the classification of the raw data signal. The raw data signal is compressed in accordance with the compression scheme.