Sensor Network Calibration via Proxy Reference Data Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Air quality regulatory monitors require frequent and costly calibration, making them economically infeasible for low-cost environmental sensor networks, especially in high spatial density networks, due to issues like missing data, high fieldwork costs, and inability to adapt to calibration drift.
Innovation Solution
A new calibration method that adjusts the gain and offset of sensor measurements to match the probability distribution of a proxy reference, combining data from regulatory stations, satellite instruments, and computer models, allowing for real-time updates without assuming sensor stability or relying on training data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If periodic co-location with reference monitor is used for calibration, then calibration accuracy is improved, but operational cost and fieldwork requirements increase significantly
Solution Approach 1:
The patent creates a virtual reference monitor by aggregating and weighting data from multiple sources (regulatory monitors, satellite instruments, computer models) to generate proxy reference measurements. This virtual copy replaces the need for physical co-location with a single expensive reference monitor, significantly reducing fieldwork costs while maintaining calibration accuracy through diversified data sources.
Solution Approach 2:
The patent combines multiple data sources including regulatory air quality station data, satellite-based instrument data, and computer model predictions into a unified proxy reference. This merging of diverse sources creates a more robust reference system than any single source could provide, improving calibration reliability while eliminating the need for expensive physical co-location.
2Productivity
If mobile reference monitor visits each sensor for calibration, then calibration coverage is improved, but reliability decreases due to insufficient pollutant variation
Solution Approach 1:
The patent merges data from multiple fixed regulatory monitors across the network to create a unified proxy reference. This aggregation ensures sufficient pollutant variation is captured across the entire network, eliminating the reliability problem of mobile references that may visit locations during stagnant atmospheric conditions. The combined data from multiple geographic locations guarantees adequate variation for reliable calibration.
Solution Approach 2:
The patent transitions from temporal calibration (mobile reference visiting sensors at different times) to spatial calibration (using data from multiple fixed regulatory monitors across different locations). This dimensional shift from time-based to space-based calibration ensures sufficient pollutant variation is captured through geographic diversity rather than temporal sampling, improving reliability while maintaining comprehensive coverage.
3Device complexity
If blind calibration using cross-correlations is used, then ground truth data requirements are reduced, but robustness against missing data decreases
Solution Approach 1:
The patent combines multiple data sources with different temporal and spatial characteristics to create the proxy reference. This diversification means that if one data source has missing data, other sources can compensate, providing robustness against data gaps. The weighted combination of regulatory monitors, satellite data, and model predictions ensures continuous calibration capability even when individual sources have interruptions.
Solution Approach 2:
The patent dynamically adjusts the weighting parameters of different data sources in the proxy reference based on data quality and availability. When certain sources have missing or low-quality data, the system automatically changes the weights to rely more on available sources, maintaining calibration robustness without requiring perfect data from all sources. This adaptive parameter adjustment provides flexibility in handling data gaps.
4Ease of manufacture
If data fusion with atmospheric model is used, then operational cost is reduced, but traceability and sensor stability assumptions are compromised
Solution Approach 1:
The patent introduces a proxy reference as an intermediary between raw sensor data and calibrated results. This proxy reference, constructed from weighted combinations of traceable sources (regulatory monitors with certified calibrations, satellite instruments, and validated computer models), provides the traceability chain that pure atmospheric models lack. The proxy reference acts as a mediator that maintains traceability while enabling low-cost calibration of sensor networks.
Data Source
AI summary
Multiple low cost individual sensors communicate with a server. A proxy sensor communicates with the server. The server periodically compares information from the individual sensors with information from the proxy sensor. The server validates and accepts information from individual sensors. The server changes gain and offset values for individual sensors providing information that is not validated by the comparing.