Optimization of Traffic and Time Control with Sensor-Driven Transmission Control System using MANET and Machine Learning

In the dynamic landscape of modern urban environments, the optimization of transportation and traffic management systems is paramount for ensuring efficiency, sustainability, and the overall well-being of inhabitants. This study introduces an innovative solution through the deployment of a sophisticated network of strategically positioned sensors at crucial locations, encompassing roadway sections, entry and exit points, and parking spaces within a smart city framework. These sensors continuously capture real-time data on traffic flow, vehicle counts, and parking spot occupancy, forming the foundational elements of an intelligent system. Robust machine learning models, notably Convolutional Neural Networks (CNN), VGG19, k-Nearest Neighbors (KNN), and Support Vector Machines (SVM), are employed to process and analyze the acquired sensor data. CNN, renowned for its prowess in image recognition, emerges as the leading performer, achieving an impressive accuracy of 97.6%. This model excels in deciphering complex spatial patterns, significantly enhancing the precision of traffic forecasts. Additionally, performance metrics, encompassing precision, recall, F1 score, and accuracy, underscore the efficacy of each machine learning model. The exceptional precision and recall values across all models showcase their robustness in identifying and predicting traffic situations, while accuracy scores affirm their overall efficiency in optimizing transportation and traffic management within smart city environments.