Distributed Sensor Network Consistency via Selective Model Updates

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Distributed sensor systems face challenges in managing communication bandwidth and computation load, particularly when a large number of sensors are used, and can fail if a central computer fails, while also struggling to ensure consistency of information across devices.

Innovation Solution

A system where devices communicate only when the information significantly improves the model, using an evaluation function that depends on the update information and current model, with devices organized in layers to reduce communication load and ensure consistency of state variables across the group.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If all sensors are coupled to a central computer that collects data from all sensors, then complete information can be obtained, but communication bandwidth and computation load become excessively high

Engineering Contradiction:
Improveinformation completenessVSAvoidcommunication bandwidth
Core Design Contradiction:
Loss of informationVSLoss of energy

Solution Approach 1:

The system divides sensors into multiple groups, with each group associated with a separate agent. Each agent maintains a local model for its group and selectively forwards information to other agents. This segmentation reduces the communication burden on any single central computer while preserving information completeness through distributed collaboration.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts only the essential information from sensor data that is necessary for model updates. Agents evaluate sensor information and selectively forward only those data points that significantly improve the model, rather than transmitting all sensor data. This extraction principle reduces communication bandwidth while maintaining information quality.

Inventive Principle:
Principle #2Taking out (Extraction)

2Ease of operation

If a central computer is used to process data from all sensors, then centralized control is achieved, but the system fails entirely when the central computer fails

Engineering Contradiction:
Improvecentralized controlVSAvoidsystem availability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The system segments the centralized control function into multiple distributed agents. Each agent independently processes sensor data from its associated group and maintains local models. This distribution eliminates the single point of failure inherent in centralized systems while preserving coordinated control through inter-agent communication.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Agents act as intermediaries between sensor groups and the overall system model. Each agent receives sensor data, processes it locally, and selectively forwards relevant information to other agents. This intermediary structure enables distributed control while maintaining system-wide consistency without requiring a central computer.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Loss of energy

If agents forward information based on local model evaluation, then communication load is reduced, but sensor information valuable to one agent may not reach that agent if another agent decides the information is not valuable

Engineering Contradiction:
Improvecommunication bandwidthVSAvoidinformation delivery
Core Design Contradiction:
Loss of energyVSLoss of information

Solution Approach 1:

The system implements feedback mechanisms where agents evaluate received information against their local models and selectively forward updates. The evaluation function determines whether information significantly improves the model, creating a feedback loop that reduces unnecessary communication while ensuring valuable information is propagated. This feedback-driven approach balances communication efficiency with information delivery reliability.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP2327231B1Method and device for operating a system with distributed sensors
Publication Date: 2017.08.09 NEDERLANDSE ORG VOOR TOEGEPAST NATUURWETENSCHAPPELIJK ONDERZOEK TNO
  • EP2327231B1 patent drawingFigure 1~2
  • EP2327231B1 patent drawingFigure 3~5
  • EP2327231B1 patent drawingFigure 4

AI summary

Distributed devices with sensors each maintain a copy of a common model. The devices each perform updates of state variables of the model in response to update information received from other devices in the group. Each device derives update information from its own sensor and decides whether or not to share the update information with the other devices. If it does not share the information the devices does not use the information itself either. Thus it is ensured that all devices keep their models consistent. Each device applies an evaluation function to the update information in the context of its current model to determine whether or not share the update information. The evaluation function may be adapted dependent on available communication bandwidth or information needs indicated by a device from a controlling layer that uses information from the model.