Air pollution is a recurring problem in large cities, which seriously affects the health of its inhabitants. Many air quality monitoring systems have been implemented with the aim of measuring the state of pollution in the city and taking actions to mitigate the effect of pollution on citizens. Cities like Medellín and its metropolitan area have robust monitoring systems (for example, SIATA), which install fixed monitoring stations in some points of the city. These stations present reliable values on air quality, but due to their high cost and high installation complexity, they cannot be deployed in large numbers, making the data only represent a few places of interest. With the emergence of new Industry 4.0 technologies, such as the Internet of Things (IoT), new low-cost monitoring systems based on IoT technologies have emerged as an option to try to solve this problem. These systems manage to capture data from many more places in the city, implementing state-of-the-art communication technologies, which allow the system to be installed in mobile agents (such as public transport buses or bicycles), achieving a greater coverage area with less installation cost. However, the quality of the data delivered by this type of system is not the best, due to the use of less robust sensors, the use of less reliable networks and the difficulty in detecting failures and correcting errors in their operation. This project proposes the development of a strategy for estimating data quality in a mobile air quality monitoring system. The proposed system will implement strategies based on Artificial Intelligence to determine the reliability of data from low-cost monitoring systems, in the future being able to determine corrective actions or adjustments to the system, tending to increase the reliability of the information collected. This system will be very useful for government entities, which will be able to make decisions knowing, a priori, the quality of the system data. Likewise, the implemented strategies will be a very important input for researchers who use pollution data to generate data processing models, such as those based on Artificial Intelligence.
GeneralImplement a strategy to determine data quality in a mobile air pollution monitoring system based on low-cost sensors.Specifics- Determine the most relevant dimensions of data quality in a mobile pollution monitoring system based on low-cost sensors.- Design the mechanisms for estimating the quality of the data for the previously determined dimensions.- Design and implement a mobile system for measuring air pollution variables.- Validate the mechanisms for estimating the quality of the data in the developed measurement system.- Carry out a feasibility analysis on the possible strategies to improve the quality of the data in the proposed system.
The research provides a solution that, based on the analysis of the quality of the data produced by low-cost sensors, would allow the air quality index to be determined more reliably. The increase in applications that make use of low-cost sensors is a global trend, so the implementation of strategies that shed light on their reliability is essential. Two (2) Articles submitted ISI WOS or Scopus; One (1) Software Registration; One (1) Bachelor's thesis from an undergraduate student; Workshop - open talk describing the project and the results obtained
|Short title||Calidad datos sensores contaminación|
|Effective start/end date||13/07/20 → 30/07/22|
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.