Optimal Extreme Learning Machine based Traffic Congestion Control System in Vehicular Network

Over the past decade, Smart cities have been advanced and minimizing traffic congestion become the main concern in the progression of smart cities. The rapid growth in the number of road vehicles has raised the number of road accidents and traffic congestion. To Solve this problem, Vehicular Network (VN) formulated several novel concepts which include traffic control, vehicular communications, and navigation. Machine Learning (ML) will be an effective technique for identifying hidden insights into ITS without being programmed clearly by learning from data. This article develops an Optimal Extreme Learning Machine based Traffic Congestion Control System (OELM-TCCS) in vehicular networks. The presented OELM-TCCS technique mainly focuses on the identification and recognition of traffic congestion in VANET. To perform this, the presented OELM-TCCS technique primarily designs a new ELM model to carry out the classification process. In addition, the presented OELM-TCCS technique executes the satin bowerbird optimization (SBO) algorithm for parameter tuning of the ELM method. To demonstrate the boosted performance of the OELM-TCCS model, a series of experiments were executed. The experimental outcomes indicate the betterment of the OELM-TCCS method over recent methods with maximum accuracy of 99.17%.