Multisensor device for emergency recognition in smart building environment

In case of serious environmental disasters, such as floods, landslides and earthquakes, the first activity that an emergency management system must carry out is the detection, recognition and report of anomalous events. In order to be able to automatically perform these actions, it is necessary to have an efficient sensor network within the building. By means of a well- organized and high-performance sensor network, data from the field are acquired and processed to produce usable information at a high level. The Building Management System (BMS) therefore has information available on the occurrence of an anomalous event, whether caused by natural or intentional causes. The purpose of our work is to design and implement a cost-effective sensorized device, managed by an ATmega microcontroller and equipped with a Wi-Fi module for easier integration, able to detect and report potentially dangerous situations within a building. In particular, fires, floods, gas leaks and earthquakes were considered. Thanks to the integration of simple data analysis and data fusion algorithms, the platform is able to recognize and report the occurrence of emergency events within a building.