Sensor Data Fusion Using Correlation Before Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor data fusion systems fail to create actionable data by correlating and fusing sensor data before storage, leading to excessive computational and storage requirements, and lack real-time data processing capabilities, especially for heterogeneous and partially heterogeneous data sources.
Innovation Solution
A system and method for sensor data fusion that includes a computer processor with curation, link, fusion, inference, and validation engines to curate and mathematically link sensor data before storage, creating a unique dataset with enhanced accuracy and reduced computational demands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If sensor data is stored before fusion, then data availability is improved, but storage requirements and computational demands increase excessively
Solution Approach 1:
The system performs preliminary correlation and fusion of sensor data before storage by mathematically linking sensor outputs in real-time. The correlation engine continuously computes correlation coefficients between sensor pairs and fuses data that exceeds threshold values, storing only the fused results rather than raw sensor data. This preliminary action maintains data availability while dramatically reducing storage requirements.
Solution Approach 2:
The system extracts only the essential correlated information from sensor data through mathematical correlation analysis. By identifying and extracting only those data pairs that exhibit significant correlation (above threshold values), the system separates useful information from redundant data, storing only the extracted correlated relationships rather than complete sensor datasets.
2Productivity
If sensor data fusion is performed in real-time, then actionable data is provided timely, but computational processing requirements increase
Solution Approach 1:
The system applies local quality by focusing computational resources only on sensor pairs that exhibit significant correlation. Rather than uniformly processing all sensor combinations, the correlation engine identifies specific sensor pairs with correlation coefficients exceeding thresholds and performs fusion only on those localized relationships. This selective approach enables real-time processing while reducing overall computational burden.
Solution Approach 2:
The system dynamically adjusts correlation thresholds based on operational conditions and data characteristics. By changing the threshold parameter adaptively, the system optimizes the balance between real-time processing capability and computational requirements, processing more data pairs when thresholds are lower and fewer when thresholds are higher, thus managing computational load while maintaining real-time functionality.
3Adaptability or versatility
If heterogeneous sensor data is fused, then data comprehensiveness is improved, but data correlation and fusion difficulty increase
Solution Approach 1:
The system employs universal mathematical correlation methods that can handle multiple sensor types and data formats uniformly. The correlation engine uses standardized statistical correlation coefficients that work across heterogeneous sensors (accelerometers, gyroscopes, magnetometers, barometers, GPS), providing a universal approach to measuring relationships between different physical quantities despite their diverse units and characteristics.
Data Source
AI summary
Systems and methods of sensor data fusion including sensor data capture, curation, linking, fusion, inference, and validation. The systems and methods described herein reduce computational demand and processing time by curating data and calculating conditional entropy. The system is operable to fuse data from a plurality of sensor types. A computer processor optionally stores fused sensor data that the system validates above a mathematical threshold.