AI-Driven Intelligent Transportation Systems in the Age of 5G/6G Networks

Abstract

A multitude of challenges confront Intelligent Transport Systems (ITS) due to the rapid growth in demand for wireless connectivity, the more diverse and het- erogeneous nature of 5G network traffic, and the likelihood of 6G being even more complicated. This research offers an AI-driven approach created especially for ITS in the context of 5G/6G networks to overcome these challenges. The goal of the research is to create a customised prediction model for traffic forecasting in ITS by analysing the efficiency of the 5G/6G network. Our proposed model, known as Refinished Long Short-Term Memory (RLSTM), employs AI methods to produce precise predictions. It dynamically adjusts hidden units and layers for enhanced accuracy. To tackle 5G traffic’s unpredictability, the model utilizes seasonal time differences to stabilize the output sequence from the original time series. Exper- imental analyses show that the RLSTM algorithm significantly improves the 5G traffic prediction performance when compared to conventional methods. This AI-driven concept offers a potentially effective way to address problems with Intelligent Transport Systems in the context of 5G networks. Its capacity for providing precise predictions enables better decision-making for efficient traffic management.