Intelligent Prediction and Scheduling of Network Traffic based on Artificial Intelligence

Abstract—With the rapid development of communication technology, network traffic prediction and scheduling become the key to improve network efficiency and user experience. In this paper, we propose an artificial intelligence-based network traffic prediction framework that aims to reduce the computational and storage burden on cloud servers. The framework utilises the MAML algorithm to quickly adapt to new tasks, especially when the new task has only a small amount of data, and learns the initial parameters of a model that can be quickly adapted on the new task. This paper also proposes three lightweight prediction models for edge devices: 1D-CNN, Self-Attention, and CNN-LSTM, which can efficiently handle temporal data and reduce the computational burden while maintaining a good prediction performance. Experimental results show that the MAML method outperforms the traditional baseline method in terms of prediction accuracy, especially when dealing with "in_traffic" and "out_traffic" features. In addition, this paper discusses the application of AI techniques in communication networks, including the optimisation of wireless and wired communication networks, and points out the limitations of the current research and the future development direction