HARDWARE DESIGN AND IMPLEMENTATION FOR BLUETOOTH LOCALIZATION BASED SOCIAL DISTANCING VIOLATION DETECTION

• Main Author: Husni, Faizal
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/55934
• Institution: Institut Teknologi Bandung

The COVID-19 pandemic has a major impact on all sectors of the world. The disease, which first appeared at the end of 2019, is highly contagious. COVID-19 is transmitted through the air, so transmission will get severe if there are large gatherings or lots of interactions. To overcome the transmission of COVID-19, many regulations have been made in the form of prohibitions to gather with lots of people and implementing social distancing. Social distancing is the practice of keeping distance from one person to another with a minimum distance of 1.8 meters. Even though these regulations have been implemented, many people still violate these regulations. Therefore, we need a system that can detect social distancing violations. In this final project, a system, that can detect social distancing violations by using a location estimation method using a Bluetooth Low Energy (BLE) device, is developed. Location estimation is carried out using LSTM model with an input in the form of a BLE signal strength. The results of the estimated location will be displayed on the web application. ESP32 is used as BLE hardware for transmitter and receiver components. BLE hardware is responsible for providing data in the form of RSSI readings to web applications. This data will then be used by the machine learning model to estimate location. The BLE device developed has succeeded in providing real-time data to web applications. The data generation process has also been carried out to produce training data for machine learning model. The data generation has an error of 3,403 dB when compared to real data. This shows that the data generation process has succeeded in generating data that describes real data.