Intelligent 5G Vehicular Networks: An Integration of DSRC and mmWave Communications

Abstract

5G heterogeneous wireless networks are believed to be an emerging and promising solution to meet growing demands posed by various vehicular telematics and infotainment applications. However, there are some challenges in designing a feasible and efficient 5G heterogeneous wireless network for pervasive vehicular connectivity, such as the complexity in the large-scale heterogeneous architecture and the dynamic nature of vehicular applications. In this paper, we propose an intelligent 5G heterogeneous wireless network architecture which includes a reinforcement learning (Q-learning)-based Dedicated Short Range Communication (DSRC) for Vehicle-to-Vehicle (V2V) and millimeter wave (mmWave) for Vehicle-to-Infrastructure (V2I) communications. Based on our findings, the goal is to develop essential vehicular connectivity management components and a situationally-aware central intelligent wireless optimizing unit to coordinate network selection for optimal vertical handoff decisions among these heterogeneous access radios. Performance evaluation supplemented by large scale simulation and field test are also provided to show our efforts in delivering our commitment.