Title: Federated Traffic Synthesizing and Classification Using Generative Adversarial Networks

Abstract: With the fast growing demand on new services and applications as well as the increasing awareness of data protection, traditional centralized traffic classification approaches are facing unprecedented challenges. This paper introduces a novel framework, Federated Generative Adversarial Networks and Automatic Classification (FGAN-AC), which integrates decentralized data synthesizing with traffic classification. FGAN-AC is able to synthesize and classify multiple types of service data traffic from decentralized local datasets without requiring a large volume of manually labeled dataset or causing any data leakage. Two types of data synthesizing approaches have been proposed and compared: computation-efficient FGAN (FGAN-\uppercase\expandafter{\romannumeral1}) and communication-efficient FGAN (FGAN-\uppercase\expandafter{\romannumeral2}). The former only implements a single CNN model for processing each local dataset and the later only requires coordination of intermediate model training parameters. An automatic data classification and model updating framework has been proposed to automatically identify unknown traffic from the synthesized data samples and create new pseudo-labels for model training. Numerical results show that our proposed framework has the ability to synthesize highly mixed service data traffic and can significantly improve the traffic classification performance compared to existing solutions.