Sensor Fusion via Radio Frame and Image Attribute Association

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing sensor fusion systems face challenges in accurately associating objects sensed by multiple sensors due to the reliance on location estimation accuracy, leading to potential wrong associations or delayed association processes, especially when multiple objects are present in a given area.

Innovation Solution

The system analyzes radio frames and images from multiple sensors to extract attribute information, using both location and attribute data to judge associations between objects, thereby improving the reliability and speed of object association.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If association is judged based only on location estimation information, then the association process is simple, but the reliability of association decreases when multiple objects are present in a given area

Engineering Contradiction:
Improveassociation reliabilityVSAvoidinformation processing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent combines multiple types of information (location estimation information and attribute information from radio frame analysis) to judge associations between objects. By merging these different information sources, the system achieves higher association reliability without excessively increasing processing complexity, as the attribute information helps distinguish between multiple objects in the same location area.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent changes the parameters used for association judgment from solely location-based to include both location and attribute parameters. By analyzing radio frame attributes (such as communication patterns, device types, or signal characteristics) in addition to location, the system can reliably distinguish between multiple objects even when they are close together, thereby improving association reliability.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If multiple types of sensor information are integrated for association, then the accuracy of association improves, but the time required for association increases

Engineering Contradiction:
Improveassociation accuracyVSAvoidassociation processing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary analysis of radio frame attributes continuously, so that when association judgment is needed, the attribute information is already prepared and available. This preliminary action allows the system to quickly combine pre-analyzed attributes with current location data, reducing the time required for association while maintaining high accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The association judgment process is segmented into independent analysis components: location estimation, radio frame attribute analysis, and integrated judgment. This segmentation allows each component to process information independently and in parallel, reducing the total processing time while maintaining the benefits of multi-information integration for accurate association.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS11930476B2Sensor information integration system, sensor information integration method, program, and recording medium
Publication Date: 2024.03.12 NEC CORP
  • US11930476B2 patent drawing
  • US11930476B2 patent drawing
  • US11930476B2 patent drawing

AI summary

The present invention is designed to make it possible to draw proper associations, when integrating pieces of information estimated in individual sensors, by associating between objects sensed by a plurality of sensors. A sensor information integration system 101 includes a radio frame analyzing section 33 configured to analyze a radio frame transmitted from a first object, to extract the first attribute information, an image analyzing section 40 configured to analyze an image, in which a second object is sensed, to extract second attribute information, and an association judging section 21 configured to judge an association between the first object and the second object, using the first attribute information and the second attribute information.