Performance Analysis of ESP32-Based Area Violation Detection Using Beam, PIR, and Ultrasonic Sensors
Restricted-area security requires detection systems that can identify human intrusion rapidly and accurately while reducing false alarms caused by non-human objects. Conventional single-sensor security systems may be limited in distinguishing humans from inanimate objects crossing a monitored boundary, particularly when the system relies only on perimeter interruption or motion detection. This study aims to design and evaluate an ESP32-based area violation detection system that integrates a Beam Sensor, Passive Infrared (PIR) sensor, and HC-SR04 ultrasonic sensor using a Finite State Machine (FSM) decision model. The Beam Sensor functions as the initial perimeter trigger, the PIR sensor verifies human presence based on body infrared radiation, and the ultrasonic sensor provides supporting distance information for spatial monitoring. Sensor readings are processed through sequential FSM states to determine whether an alarm condition should be activated. The prototype was tested in 30 trials involving human and inanimate objects at distances ranging from 1 m to 5 m. The experimental results show that the proposed system achieved an accuracy of 83.33%, precision of 85.71%, recall of 80.00%, and F1-score of approximately 82.76%. The HC-SR04 ultrasonic sensor also produced an average measurement error of 1.68% within the tested range of 50–300 cm. These findings indicate that the integration of multi-sensor inputs with FSM-based validation provides a structured and selective approach for area violation detection. Although the system is designed to address limitations commonly found in single-sensor configurations, further comparative testing against beam-only, PIR-only, and ultrasonic-only systems is required to empirically confirm its relative performance advantage.