Traffic Classification
Van Tong, Sami Souihi, Hai Anh Tran, Abdelhamid Mellouk · 2023
Traffic classification plays an essential role in network troubleshooting for NOs. In this chapter, the authors present a traffic classification module to identify the application class for QUIC traffic. The application class also plays an important role in application-aware remediation approaches in network troubleshooting. This module is briefly described in the novel troubleshooting framework. The authors provide background information about convolutional neural network and the characteristics of QUIC-based applications. They also present related work on traffic classification consisting of port-based approaches, payload-based approaches, statistic-based approaches and deep learning-based methods. The authors evaluate the performance of the traffic classification method which contains two stages of classification. In the first stage, they evaluate the performance and time complexity of machine learning (ML) algorithms to select an appropriate ML algorithm. In the second stage, the authors evaluate its performance over different scenarios including different subsets of the input vector and various loss functions.