A Data Fusion Approach of Physical Variables Measured through a Wireless Sensor Network Thomas Alejandro Arias Pelayo, Guillermo Molero Castillo, Everardo Bárcenas, Rocío Aldeco Pérez Universidad Nacional Autónoma de México, Facultad de Ingeniería, Mexico thomas.arias95@gmail.com, {gmoleroca, raldeco}@fib.unam.mx, ebarcenas@unam.mx Abstract. At present, current data fusion methods are a useful tool for integrating data sources, prior to data analytics, and provide a unified view of an observed phenomenon or event. This paper presents the development of the Monte Carlo method, as a data fusion mechanism, obtained from a wireless sensor network. This network of sensors was designed and installed in a closed environment of human occupation. The data collected was of physical variables, such as temperature, humidity, and dust density, which were stored in the cloud through ThingSpeak, which is an open-source platform for the Internet of Things. As a result, it was succeeded in data fused properly and the method was evaluated through the root-mean-square error. Undoubtedly, fused values can be useful, for example, for the analysis of the thermal comfort of users in closed environments, where there are minimal ventilation rates and adequate indoor air quality is needed. Keywords: Data Fusion, Monte Carlo Method, Wireless Sensor Network, Physical Variables. 1 Introduction Today, the automation of industrial processes, the improvement of workforce capabilities and the development of new products through Artificial Intelligence benefits society as part of the digital transformation process in this era of the fourth industrial Revolution. However, of all the technologies to consider for their potential, without a doubt, wireless sensor networks (WSN) are proving their usefulness for measuring and storing data on a specific environment, as a key technology to implement the Internet of Things (IoT), whose main characteristic is its low energy consumption and its deployment in inaccessible locations or even integrated within structures. It is important to note that currently, wireless sensor networks are autonomous devices that work collaboratively, distributed throughout an area of interest and 103 ISSN 1870-4069 Research in Computing Science 149(11), 2020 pp. 103–113; rec. 21-08-2020; acc. 19-10-2020