A Survey of Encrypted Traffic Classification: Datasets, Representation, Approaches and Future Thinking

Traffic classification technologies have evolved as a key component for modern network protection and management systems. However, encryption algorithms have been adopted in traffic transmission to enhance communication security and privacy protection. Traditional traffic classification methods that rely on valuable information from payloads gradually become ineffective. To this end, researchers have conducted a lot on encrypted traffic classification. In this paper, we present a comprehensive survey on recent achievements in traditional-machine-learning (TML) or deep-learning (DL) powered encrypted traffic analysis. To begin with, we review the literature in this area and summarize the analysis goals. Secondly, we abstract the overall framework of encrypted traffic classification, including traffic collection, representation, analysis methods, and performance evaluation. Then a comprehensive review of existing research was conducted from these four aspects. Lastly, we conclude with challenges and directions for future research in encrypted traffic classification.