Sistema de detecção de queda e monitoramento da frequência cardíaca utilizando ESP8266 e Protocolo MQTT

Abstract

According to the IBGE there is a continuous increase in the number of single-person arrangements. The main causes for this, are decline in fertility and aging of population. With the increase in the number of people living alone comes the problem of who will help in emergencies case. This problem become more alarmant by the fact that brazilian population is undergoing an aging process. One solution to this is people monitoring systems. However, the development of a monitoring system is not a trivial task because it involves several different data sources. Besides that, it is necessary to transform data into information for later presentation and provide a way to manage the entire solution. The idea of this work is to present a simple architecture and economically accessible system for fall detection and heart rate monitoring. The goal is to develop a solution to people who need to be monitored by relatives or a responsible one, without interference of companies, hospitals, clinics, etc. The proposed system aims to be as an extra security for people who live alone, elderly or not. In order to solve some of the problems mentioned above, this work was guided by the concept and solutions present in Internet of Things (IoT). It was used an accelerometer to detect falls and a sensor to monitor heart’s electrical signals. These two sensors were connected to Wi-Fi module, ESP8266, which is responsible for preprocessing data, generating alerts and sending them to broker with MQTT communication protocol. A web system is the interface for information management and presents detected alerts of all sensors connected to ESP8266. Finally, a mobile application allows to view detected alerts of authenticated user as well as the users for whom he is responsible. According to the performance tests in web application and broker, it was verified that it is possible to send 1500 simultaneous alerts without failures. The ESP8266 board proved suitable presenting a battery life of almost 8 days without disconnecting the equipment. The fall detection algorithm was also effective, with 93% accuracy.