Prototipe Sistem Monitoring dan Peringatan Dini Bahaya Maling dan Kebakaran Berbasis Internet of Things (IoT)

Abstract

This study aims to design and build a prototype of an Internet of Things (IoT)-based monitoring and early warning system for burglary and fire as an alternative to home security systems that use Passive Infrared (PIR) sensors. The system integrates two ESP32 units that communicate via UART. Unit 1 uses three beam-interruption-based laser/LDR sensors to detect intrusions, an MQ-2 gas sensor to detect gas leaks, and a flame sensor to detect fire, and is equipped with an LCD and a buzzer. Unit 2 uses an ESP32-CAM to send image notifications via Telegram, upload data to ThingSpeak, and provide live camera streaming. The system was developed using the Waterfall model, which includes requirements analysis, design, implementation, and testing. Test results show that the laser/LDR sensors can classify conditions as normal, warning, animal detected, and intruder with a one-second debounce. The MQ-2 sensor can distinguish between normal, alert, and danger conditions using a moving average and hysteresis, while the flame sensor detects fire with a 300-millisecond debounce. UART communication, Telegram notifications, ThingSpeak visualization, and live streaming all functioned properly. Thus, the prototype is capable of providing more targeted intrusion detection and early warnings of fire hazards.