High Density Real-Time Air Quality Derived Services from IoT Networks

Area of Science:

• Environmental Science
• Computer Science
• Data Science

Background:

• Growing demand for user-centric air quality services requires high-density data.
• Existing scattered sensor data presents challenges for scalable, real-time applications.
• Need for robust data interpolation and anomaly detection for reliable air quality monitoring.

Purpose of the Study:

• To develop and optimize data interpolation algorithms for generating dense, regular air quality data from scattered sources.
• To create a scalable, on-demand service for providing air quality data across urban areas.
• To leverage interpolation error analysis for detecting sensor dysfunctions and enabling proactive maintenance.

Main Methods:

• Evaluation of various interpolation algorithms for accuracy against real-world air quality data.
• Development of a Micro Services-based data analytics solution using Node.JS Node-RED (IoT Applications).
• Integration of the Snap4City platform for real-time air quality data processing and service delivery.

Main Results:

• Optimized interpolation algorithms demonstrated effectiveness in creating dense air quality datasets.
• The system successfully provided on-demand air quality data and heatmaps for urban environments.
• Interpolation error analysis proved effective in identifying sensor malfunctions, aiding maintenance efforts.

Conclusions:

• The integrated solution offers a scalable approach to delivering high-density air quality data services.
• The developed methods enhance the reliability and utility of air quality monitoring systems.
• The Snap4City platform, applied in Helsinki, shows promise for widespread adoption in European cities.