TFE-GNN: A Temporal Fusion Encoder Using Graph Neural Networks for Fine-grained Encrypted Traffic Classification

TL;DR

This paper introduces TFE-GNN, a novel graph neural network-based model that captures fine-grained features of encrypted traffic at the byte level, outperforming existing methods in classification accuracy.

Contribution

It proposes a byte-level traffic graph construction method and a dual embedding, GNN-based encoder with feature fusion for improved encrypted traffic classification.

Findings

TFE-GNN outperforms state-of-the-art methods on real datasets.

The model effectively captures byte-level correlations.

Fusion of header and payload enhances classification accuracy.

Abstract

Encrypted traffic classification is receiving widespread attention from researchers and industrial companies. However, the existing methods only extract flow-level features, failing to handle short flows because of unreliable statistical properties, or treat the header and payload equally, failing to mine the potential correlation between bytes. Therefore, in this paper, we propose a byte-level traffic graph construction approach based on point-wise mutual information (PMI), and a model named Temporal Fusion Encoder using Graph Neural Networks (TFE-GNN) for feature extraction. In particular, we design a dual embedding layer, a GNN-based traffic graph encoder as well as a cross-gated feature fusion mechanism, which can first embed the header and payload bytes separately and then fuses them together to obtain a stronger feature representation. The experimental results on two real datasets demonstrate that TFE-GNN outperforms multiple state-of-the-art methods in fine-grained encrypted traffic classification tasks.