1084 IEEE SENSORS JOURNAL, VOL. 20, NO. 2, JANUARY 15, 2020 PM 2.5 Concentration Measurement Analysis by Using Non-Parametric Statistical Inference Wilmar Hernandez , Senior Member, IEEE , Alfredo Mendez, Angela Maria Diaz-Marquez, and Rasa Zalakeviciute Abstract —Rapid population growth, urbanization and motorization have brought about secondary effects that have gradually damaged the atmosphere, whose importance is vital for both the survival of all living beings and the climate bal- ance. In this sense, air pollution is a problem that affects cur- rent society and is much more critical in developing countries. In this context, in the present paper non-parametric statistical inference techniques are used to carry out the analysis of measurements of health concerning fine particulate matter concentration, PM 2.5 , in an urban park of Quito, Ecuador. In short, the data collected during the measurements were stored in random variables and the Kruskal-Wallis test was used to test if these random variables come from populations with identical distributions. Also, the Wilcoxon signed rank test was used to test if the numerical values collected in the samples of the random variables of interest represent a level of contamination that could be dangerous for human beings. The experimental results show that urban parks and, specifically, trees are a natural filter between the pollution generated in the road and the center of the park. Therefore, the role of trees in the face of vehicular pollution will depend on two variables: the amount and compactness of the vegetation, and the emission levels recorded in the border roads. Index Terms— PM 2.5 concentration, air pollution, non-parametric statistical inference, Kruskal-Wallis test, Wilcoxon signed-rank test. I. I NTRODUCTION A NTHROPOGENIC burden on the environment originates from the overpopulation of the world. As a result of the rapid increase in human population and subsequent anthro- pogenic activities, we face deteriorating air quality, which is the largest single environmental health risk [1]. Another Manuscript received August 15, 2019; revised September 12, 2019; accepted September 30, 2019. Date of publication October 4, 2019; date of current version December 31, 2019. This work was supported in part by the CEDIA-Ecuador, under Project CEPRA XII-2018-13, in part by the Universidad de Las Américas, Quito, Ecuador, under Project ERa.ERI.WHP.18.01, and in part by the Universidad Politécnica de Madrid, Spain. The associate editor coordinating the review of this article and approving it for publication was Dr. Shanhong Xia. (Corresponding author: Wilmar Hernandez.) W. Hernandez is with the Facultad de Ingeniería y Ciencias Aplicadas, Universidad de Las Américas, Campus Queri, Quito 170504, Ecuador (e-mail: wilmar.hernandez@udla.edu.ec). A. Mendez is with the Departamento de Matemática Aplicada a las Tecnologías de la Información y Comunicaciones, ETS de Ingenieros de Telecomunicación, Universidad Politécnica de Madrid, 28040 Madrid, Spain (e-mail: alfredo.mendez@upm.es). A. M. Diaz-Marquez is with the Grupo Dinámicas + Lugar, Medio y Sociedad (DL+LMS), Universidad de Las Américas, Campus Queri, Quito 170504, Ecuador (e-mail: angela.diaz@udla.edu.ec). R. Zalakeviciute is with the Grupo de Biodiversidad, Medio Ambiente y Salud (BIOMAS), Universidad de Las Américas, Campus Queri, Quito 170504, Ecuador (e-mail: rasa.zalakeviciute@udla.edu.ec). Digital Object Identifier 10.1109/JSEN.2019.2945581 global trend is urbanization, nevertheless, it is different in different parts of the world, with Americas being the most urbanized continents [2]. As expected, air pollution problems are quite common in urban areas, with most of the cities (i.e., population over 100000) exceeding air quality stan- dards, and these problems are especially critical in developing countries [3]. Apart from natural air pollution, such as volcano eruptions, wildfires, and dust storms, most of the toxic atmospheric contamination comes from human activities, such as industries, agriculture and traffic. In developing countries, traffic is one of the major sources of health concerning fine particulate matter, PM 2.5 , which are particles with an aerodynamic diameter less than 2.5 μm [4]. An increasing number of studies show that PM 2.5 causes asthma, an array of respiratory problems and inflammation, jeopardizes lung functions and even promotes cancer [5]–[7]. Besides, there is a growing amount of scientific evidence demonstrating a relationship between the proximity to traffic and cardiopulmonary mortality [8]. Therefore, the exposure to these toxic components should be avoided or minimized. Due to the importance of studying the PM 2.5 , in the last years there have been numerous research initiatives to design more economic sensors to measure this physical quantity and to estimate of its magnitude, especially important in urban This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see http://creativecommons.org/licenses/by/4.0/