Wireless Sensor Network Combined with Cloud Computing for Air Quality Monitoring

Area of Science:

• Environmental Science
• Sensor Technology
• Artificial Intelligence

Background:

• Traditional air quality monitoring systems often lack sufficient spatial resolution.
• Emerging low-cost, densely distributed wireless sensors offer improved ambient air quality data.
• Need for efficient, cost-effective methods for monitoring air pollutants.

Purpose of the Study:

• To present a novel wireless sensor network (WSN) for air quality measurement.
• To demonstrate a cloud-based system for data processing and analysis using AI.
• To validate the system's performance in detecting specific volatile organic compounds.

Main Methods:

• Deployment of a distributed network of low-power ZigBee sensor nodes.
• Data transmission via a gateway to an optimized cloud computing system.
• Application of AI techniques (PCA, MLP, SVM) for data processing and compound detection.

Main Results:

• The WSN system effectively stores, monitors, processes, and visualizes air quality data.
• AI techniques demonstrated good performance in discriminating and quantifying volatile organic compounds.
• Laboratory case study confirmed system applicability for detecting benzene, toluene, ethylbenzene, and xylene.

Conclusions:

• The proposed low-cost, low-size, low-power WSN significantly enhances air quality measurement efficiency.
• Cloud-based AI processing enables accurate detection and quantification of air pollutants.
• The system provides valuable spatial information for air quality distribution assessment.