Causal Network Dataflow Control for IoT Sensor Networks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In IoT environments, it is challenging for developers to design applications that effectively collect and utilize sensing data from a vast network of sensors due to the complexity of causal relationships between various sensors and the dynamic nature of sensor availability, leading to inefficiencies in data distribution and utilization.

Innovation Solution

A dataflow control apparatus that utilizes a causal network to identify and select devices with causal relationships to an objective index, providing data that is automatically relevant to the application's objectives, even in dynamic network conditions, by employing transfer entropy to evaluate causal strength and time delays.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If all sensors are connected to the network to enable comprehensive data collection, then the availability of sensing data increases, but the network congestion and communication overhead increase

Engineering Contradiction:
Improveavailability of sensing dataVSAvoidnetwork congestion
Core Design Contradiction:
Quantity of substanceVSLoss of energy

Solution Approach 1:

The system pre-establishes a causal network that maps relationships between sensors and application objectives before data collection begins. This preliminary structuring enables the system to quickly identify and retrieve only the relevant data needed for specific applications, avoiding the need to transmit or process all available sensor data through the network.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system extracts and transmits only the specific sensing data that has a causal relationship with the application objective, rather than transmitting all available sensor data. This selective extraction reduces network traffic and communication overhead while ensuring that the most relevant data is delivered to applications.

Inventive Principle:
Principle #2Taking out (Extraction)

2Loss of information

If developers manually define app-side metadata to specify required sensing data, then data relevance to application objectives improves, but the complexity of application design and development increases

Engineering Contradiction:
Improvedata relevanceVSAvoidapplication design complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The system automatically performs the task of identifying and selecting relevant sensing data by computing causal relationships between sensors and application objectives. Instead of requiring developers to manually specify data requirements, the system self-determines which sensors are relevant based on the objective, thereby reducing development complexity while maintaining data relevance.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The causal network computation acts as an intermediary between the application objective and the sensor network. This intermediary automatically translates high-level application objectives into specific sensor selections, eliminating the need for developers to directly manage the complex mapping between sensors and application requirements.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Stability of the object's composition

If the system uses traditional metadata matching to control dataflow, then the data distribution follows predefined rules, but the system cannot adapt to dynamic sensor availability and network conditions

Engineering Contradiction:
Improvedata distribution stabilityVSAvoidadaptability to dynamic conditions
Core Design Contradiction:
Stability of the object's compositionVSAdaptability or versatility

Solution Approach 1:

The system dynamically computes causal relationships between sensors and application objectives based on current network conditions and sensor availability. Rather than relying on static, pre-defined data distribution rules, the system continuously adapts its data selection and routing decisions to reflect the current state of the network and application needs.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system incorporates feedback mechanisms that monitor sensor availability and network conditions, using this information to adjust dataflow control decisions in real-time. This feedback loop enables the system to maintain stable and relevant data distribution even as underlying conditions change, by continuously re-evaluating causal relationships and adapting accordingly.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3318976B1Data flow control device and data flow control method
Publication Date: 2023.11.29 OMRON CORP
  • EP3318976B1 patent drawingFigure 1
  • EP3318976B1 patent drawingFigure 2
  • EP3318976B1 patent drawingFigure 3

AI summary

A dataflow control apparatus has a causal network storage unit that stores a causal network representing a causal relationship between a plurality of indices that include at least a plurality of device indices which are respectively data obtained in a plurality of devices and an objective index which is an index that an application controls or predicts, a device index selection unit that selects, based on the causal network, at least one device index having a causal relationship with the objective index of the application, from among the plurality of device indices, in a case where a data request is received from the application, and a dataflow control unit that controls the dataflow, such that data obtained in the device corresponding to the selected device index is provided to the application.