ESP32-Based Wi-Fi CSI for Traffic Monitoring and Congestion Detection: Enabling Signal Automation for Efficient Urban Transportation

Traffic congestion has become a key problem in cities, which affects travel times, fuel consumption, and ultimately pollution. Current congestion monitoring and management approaches are costly, intrusive, and not scalable. In this paper, we proposed an innovative and low-cost solution for real-time monitoring of vehicle traffic and congestion using Channel State Information (CSI) from Wi-Fi signals. The primary goal of this research is to develop an Internet of Things (IoT)-based system, using an ESP32 microcontroller to capture CSI data, and then use machine learning models on the data to determine traffic congestion. This study applies Artificial Neural Networks (ANN) to classify vehicle density and, consequently, manage the timing of traffic light signals based on real-time congestion information. The system was tested in a variety of traffic scenarios consisting of six types of vehicles with accuracy 89%. The results of the study showed that the CSI-based approach was better than the current methods in use with high precision and low cost. The study determined that the use of CSI data combined with machine learning provided a low-cost, scalable, and non-intrusive solution to managing traffic congestion and could greatly improve urban traffic and congestion levels.