An IoT-Enabled E-Nose for Remote Detection and Monitoring of Airborne Pollution Hazards Using LoRa Network Protocol

Area of Science:

• Environmental Science
• Sensor Technology
• Artificial Intelligence

Background:

• Electronic noses (e-noses) are vital for detecting airborne hazards and preventing accidents.
• Current e-nose networks often rely on power-intensive internet connectivity.
• Long-range (LoRa) wireless networks offer a low-power alternative for remote monitoring.

Purpose of the Study:

• To propose and develop a networked intelligent gas sensor system (N-IGSS) for real-time airborne pollution hazard detection and monitoring.
• To utilize LoRa low-power wide-area networking for independent operation without internet connectivity.
• To evaluate the system's performance in identifying various airborne hazards.

Main Methods:

• Developed a gas sensor node with seven cross-selective metal-oxide semiconductor (MOX) gas sensor elements, a microcontroller, and a LoRa module.
• Exposed the sensor node to six classes: ambient air, five volatile organic compounds (VOCs), and emissions from burning tobacco, paints, carpets, alcohol, and incense.
• Applied a two-stage analysis space transformation: standardized linear discriminant analysis (SLDA) for preprocessing and four classifiers (AdaBoost, XGBoost, RF, MLP) for training and testing.

Main Results:

• The N-IGSS achieved 100% accuracy in identifying 30 unknown test samples.
• A low mean squared error (MSE) of 1.42 × 10^-4 was recorded.
• Successful detection and monitoring were achieved over a distance of 590 meters.

Conclusions:

• The proposed N-IGSS effectively detects and monitors airborne hazards in real-time using LoRa technology.
• The system demonstrates high accuracy and low error rates for pollutant identification.
• N-IGSS offers a viable, low-power solution for widespread airborne hazard monitoring in remote locations.