IoT Based Smart Parking Prediction System

With rapid urbanization, finding available parking spaces in congested areas has become a
significant challenge for vehicle owners worldwide. Conventional parking systems lack real-time
availability information, resulting in wasted time, increased fuel consumption, and heightened traffic
congestion. This paper proposes an IoT-Based Smart Parking Prediction System that leverages HCSR04
ultrasonic sensors, NodeMCU ESP8266 microcontrollers, and a Random Forest machine learning
algorithm to detect, monitor, and predict parking space availability in real time. Sensor data is
transmitted to an AWS EC2 cloud server via the MQTT protocol over Wi-Fi, and slot information is
presented to users through a ReactJS web dashboard and mobile interface.