The need for vehicles to communicate with each other in real-time using high-speed networks have a problem maintaining Quality of Service (QoS). This is because the network's constantly changing conditions, topology, and handovers. This paper proposes a machine learning based context-aware system designed for vehicle networks with high mobility. The system is designed to predict and estimate context with optimization methods. This allows the system to optimally allocate resources, transmit sensitive data, and change settings dynamically. In our experiments, mobility patterns were simulated, and the system was tested on throughput, latency, and packet loss. The results were consistent with our hypothesis. The system demonstrated a 32% increase in throughput and a 27% decrease in latency during the high-speed tests. The system is also able to keep adapting during network changes, this positively affects consistency and reliability. This serves as a step forward towards dependable communication systems for smart transport systems, infrastructure, and vehicles.
Tripathy et al. (Mon,) studied this question.