An Architecture for the Internet of Things and the Use of Big Data Techniques in the Analysis of Carbon Monoxide

Abstract

The use of sensors for the monitoring of a given environment allied to the Internet as a means of communication is popularly known as Internet of Things (IoT). The amount of information generated in this environment has led to an unprecedented increase in data collection. One of the major challenges for its development lies in the storage and the processing of this huge volume of data into acceptable measurement and analysis parameters. This research takes up this challenge by storing and compiling data from different sensors, and by carrying out an exploratory analysis of the information gathered. In this research, sensors that collect data from a specific Sao Paulo's Metropolitan Area (SMA) have been analysed. These sensors are capable of measuring carbon monoxide (CO) levels. This research aims to analyse the main architectures for both batch and stream sensor processing and to use one of them for the construction of a Big Data environment. Big Data tools were used for IoT storage, processing and visualization data. During the experiments, carbon monoxide sensors (MQ7), were analysed. They were connected through a microcontroller unit that supports the Transmission Control Protocol/Internet Protocol (TCP/IP). This project highlights the necessary tools to execute and analyse the data in a dynamic manner. The data collected by the sensors show that the avarage levels of carbon monoxide are well above the international standards set by the World Health Organization (WHO).