ABSTRACT Air pollution is one of the major health problems in the world, especially in urban regions. Therefore, air quality monitoring is crucial to protect public health. Existing IoT‐based solutions are often limited by high cost, poor sensor accuracy, or a lack of real‐time accessibility in developing countries. This paper presents a low‐cost Internet of Things (IoT)‐based solution for real‐time air quality monitoring in developing regions that focuses on the most harmful pollutants, particularly PM2.5, CO, CO 2 , and NO 2 . The system was developed using readily accessible hardware components like Arduino Uno interfaced with PMS5003 for PM2.5, MQ7 for CO, and MQ135 for CO 2 and transmits real‐time sensor data to a cloud‐based Firebase database via a Python script. It also provided access to data through a custom‐built Android app for real‐time or historic pollutant monitoring, including timely notifications for unsafe levels. The system was highly effective in tracking spikes during high‐traffic festivals, enabling timely notifications to authorities and citizens. It was tested in Cumilla, Bangladesh, where it demonstrated 99.7% operational uptime, and pollutant measurements deviated less than 5% from certified ground station data, confirming its real‐world reliability. The system was designed for both scalability and portability, supporting deployment across diverse environments and expansion for broader monitoring networks. This work is novel in demonstrating a validated, low‐cost, multi‐pollutant IoT monitoring system with mobile integration and a design emphasizing scalability, portability, and accessibility, which has not been reported in previous studies for medium‐sized South Asian cities. The work merges IoT and sustainability by offering real‐time insights into air quality, enabling public awareness and evidence‐based policymaking. This advances IoT air monitoring by combining technical accuracy, community usability, and real‐world validation, supporting smart city integration and improved urban living conditions.
Shahariar et al. (Tue,) studied this question.