Evaluation of Suitability of Low-Cost Gas Sensors for Monitoring Indoor and Outdoor Urban Areas

Air pollution is a significant environmental risk to health; reducing pollution levels could help minimize related diseases, such as cancer, asthma, and stroke. Gas sensors, such as the MQ family, can be used to establish security measures for detecting environmental problems and improving air quality. The aim of this article is to develop a low-cost system using MQ sensors and an Arduino Mega to monitor air quality in different indoor and outdoor scenarios and identify the origin of the data using different approaches such as discriminant analysis (DA) and probabilistic neural network (PNN). The system is composed of an Arduino Mega and four MQ gas sensors. The response of four different MQ sensors (MQ-2, MQ-3, MQ-7, and MQ-135) to different indoor and outdoor environments is analyzed. The results indicate that the living room and kitchen have a stable response for all sensors, while the bar and the terrace have higher variability in their response. This article presents the results of using DA and PNN to identify indoor and outdoor areas using different combinations of MQ sensors, achieving up to 99.47% correctly classified cases with all sensors using PNN. This article’s results show that their proposed system outperforms existing applications in correctly classifying cases, with well-classified cases with two sensors and the PNN reaching 98.22%.