A Lightweight Multi-Classification Model for Identifying Network Application Traffic Using Knowledge Distillation

Zhiyuan Li

doi:10.3390/fi18040197

Return

A Lightweight Multi-Classification Model for Identifying Network Application Traffic Using Knowledge Distillation

2026-04-07

Zhiyuan Li *

Yonghao Feng

DOI：10.3390/fi18040197

Original

Save

View PDF

Abstract

En 中文

To address the limitations of insufficient feature representation, large model size, and high deployment cost in network traffic classification, a lightweight classification framework based on multi-teacher knowledge distillation is proposed. The framework consists of two heterogeneous teacher networks and a compact student network to enable end-to-end traffic classification under constrained computational resources. The teacher networks incorporate complementary spatio-temporal modeling strategies, including a bidirectional temporal convolutional network (BiTCN) enhanced with attention mechanisms and convolutional neural network (CNN), and a parallel spatio-temporal fusion architecture integrating bidirectional long short-term memory (BiLSTM) and CNN. Knowledge from the teacher ensemble is distilled into a lightweight CNN-based student network through soft-target supervision, leading to improved generalization capability with significantly reduced model complexity. Experimental results demonstrate that effective knowledge transfer is achieved while reducing model parameters by more than 80%, and performance gains of about 1–3% are obtained compared with baseline methods. These results indicate strong potential for practical deployment in resource-constrained network environments.

Keywords:

network traffic identification

multi-classification

model compression

knowledge distillation

spatio-temporal feature fusion

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.