Sensor Data Fusion Using Conditional Entropy for Validated Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor data fusion systems fail to create actionable data by curating and linking sensor data before fusion, leading to excessive computational and storage requirements, and they do not generate new datasets that provide accuracy and predictive information.
Innovation Solution
A system and method for sensor data fusion that includes a computer processor with curation, link, fusion, inference, and validation engines to mathematically link and fuse heterogeneous, partially heterogeneous, or homogeneous data points, creating a unique dataset with accuracy values, reducing computational and storage demands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If sensor data is fused without prior curation and linking, then the fusion process is simpler and faster, but the computational and storage requirements become excessive and actionable data is not generated
Solution Approach 1:
The system performs curation and linking of sensor data before the actual fusion process. The curation engine pre-processes raw sensor data by filtering, organizing, and validating it, while the link engine establishes relationships between different data points. This preliminary action reduces the complexity and computational load of the subsequent fusion operation, allowing faster processing with reduced energy consumption.
Solution Approach 2:
The data fusion system is divided into distinct functional modules: a curation engine for data preparation, a link engine for establishing relationships, and a fusion engine for combining data. This segmentation allows each component to specialize in specific tasks, improving overall efficiency and reducing redundant computations, thereby lowering power consumption while maintaining high productivity.
2Measurement precision
If multiple heterogeneous sensor data points are fused to create new datasets, then measurement accuracy and predictive information improve, but device complexity increases
Solution Approach 1:
The fusion engine is designed to handle multiple types of sensor data (homogeneous and heterogeneous) using a unified mathematical framework. It can process temperature, pressure, humidity, and other sensor inputs through the same curation, linking, and fusion mechanisms. This multi-functionality reduces the need for separate processing paths for different data types, managing complexity while improving measurement precision through comprehensive data integration.
Solution Approach 2:
The system dynamically adjusts processing parameters based on the type and quality of input sensor data. The curation engine modifies data parameters such as sampling rates, filtering thresholds, and validation criteria according to sensor characteristics. This adaptive parameter adjustment optimizes the fusion process for different data scenarios, enhancing accuracy without requiring fundamentally different system architectures.
3Reliability
If all sensor data is stored for later analysis, then data availability for predictions is improved, but storage requirements become excessive
Solution Approach 1:
The system extracts only the most relevant and actionable features from raw sensor data during the curation phase. Instead of storing complete raw datasets, the curation engine identifies and extracts key parameters, trends, and patterns that are essential for predictions. This extraction reduces storage requirements while maintaining the reliability needed for accurate predictions by preserving only the most informative data elements.
Solution Approach 2:
The system performs preliminary processing and filtering of sensor data before storage, preparing it in advance for prediction tasks. The curation and link engines pre-organize data, establish relationships, and validate information beforehand, so that when predictions are needed, the system can quickly access pre-processed, high-quality data without storing unnecessary raw information, thus reducing storage volume while maintaining prediction reliability.
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.