AI Based Semantic Communication And Digital Twin Integrated Edge Architecture For 6G Vehicular Networks

Abstract

Internet of Things (IoT) applications in autonomous vehicles (AVs) have significantly driven the need for 6G-enabled vehicular networks with ultra-low latency, reliable and high bandwidth efficiency for the intelligent transportation services. But the conventional Vehicle-to-Everything (V2X) systems transmit a huge amount of raw sensor information, consuming much bandwidth, causing high communication delay, consuming much energy, and not correctly synchronizing digital twins. In order to solve the foregoing issues, a new solution combining Semantic Communication and Digital Twin is proposed for 6G Vehicular Networks, which is called as the Semantic Communication and Digital Twin Integrated Edge Architecture (SC-DTEA) for 6G Vehicular Network. Using semantic information extraction to deliver only data that is relevant to the task, while the use of edge intelligence for computational resource allocation and continuous synchronisation of the vehicular digital twin for real-time decision making. A mathematical optimization model is formulated to simultaneously minimize the latency, energy consumption, and synchronization error and maximize the reliability and spectral efficiency. The simulation results show that this approach can realize 60.9% latency reduction, 53.8% energy saving, 99.9% communication reliability, 1.79 bps/Hz spectral efficiency, 99.85% digital twin synchronization accuracy, and even 80% data reduction compared to conventional V2X approaches, which proves its effectiveness for the future intelligent transportation system of 6G.