Noise Monitoring System Development in a Library Based on The Internet of Things

A library is one of the important places for the community, especially students. However, not all visitors know the library's rules and act arbitrarily to create noise that can disturb other visitors. This research is focused on the development of classifying and monitoring the unwanted noise in the library. The system is built with the Arduino Nano 33 BLE microcontroller using the DFROBOT Analog Sound Level Meter Sense sensor and ESP32-WROOM32U. The system is equipped with classification capabilities resulting from machine training using the Convolutional Neural Network algorithm by utilizing a Feature Extraction. The system is then connected to Wi-Fi to be integrated with websites created using the PHP programming language and the Laravel framework. Data from the monitoring will be stored in the MySQL database. The system can give a noise warning when a human or cell phone sound exceeds the threshold with an average of 82.78% classification accuracy and an ideal distance from the sound source, as far as 30-100 cm.