Design and Optimization of Low-Latency Communication Protocols for 5G-Enabled Autonomous Vehicle Networks

The paper is centred on low-latency communication schemes for the multi-vehicle network in the context of 5G autonomous vehicles highlighting the issues of latency, dependability, and resource management. With the development of AV technology it is crucial to have real-time information transfer between the vehicles and the transport infrastructure for smooth and safe operation. Several novel technologies have been adopted in the design of the proposed communication protocol: Edge computing, Network slicing, Machine learning-based optimization, Cross layer design. According to edge computing, the data processing is performed near the vehicles and hence latency is minimized. This separates the network with traffic type in a manner that gives priority to the safety sensitive info over other data that is not so critical. Machine learning is always adaptive to the condition of the network by predicting the state of the network in real time. A cross-layer optimization in the protocol increases the communication between the layers at the protocol stack that also builds efficiency. The computer-based tests in different driving scenarios reveal that the latency has been reduced by more than 70%, and overall network stability has been increased by means of low packet loss rate, low jitter and fewer handover failures. Resource utilisation is done dynamically and in conjunction with edge computing to improve bandwidth utilisation, reduce CPU usage and consumed energy. Also, these findings indicate that the proposed protocol can fit the requirements of the AV networks, ensuring future developments, especially when considering the 6G systems.