A Lightweight Traffic Identification Model Based on Deep Learning

doi:10.11871/jfdc.issn.2096-742X.2025.02.001

Abstract

Abstract:

[Objective] To address the issue of deep learning-based traffic identification models typically having high parameter count and computational complexity, which limits their deployment in resource-constrained edge networks, a lightweight traffic identification model based on an improved ShuffleNetV2 is proposed. [Methods] First, appropriate resolution and width factors are selected to balance the model’s efficiency and performance. Second, a lightweight multi-scale feature fusion module and an enhanced coordinate attention mechanism module are integrated into the model to improve identification performance with minimal computational and storage overhead. Finally, the ReLU activation function is replaced with ReLU6 to enhance the model’s suitability for deployment and inference in resource-limited settings. [Results] The experiments are conducted based on the publicly available ISCXVPN2016 dataset, and the results validate the model's excellent performance in terms of recognition accuracy and execution efficiency. Compared to the widely adopted ResNet-18 model, this model achieves an accuracy increase of 0.2% (reaching 98.8%), while significantly reducing the parameter count and computational complexity by 96.46% and 62.97%, respectively. [Conclusions] The model achieves significant lightweighting while maintaining high accuracy.

Key words: traffic identification, ShuffleNetV2, attention mechanism, feature fusion

LI Yong,REN Yongmao,YIN Zhuoran,ZHOU Xu. A Lightweight Traffic Identification Model Based on Deep Learning[J]. Frontiers of Data and Computing, 2025, 7(2): 3-11, https://cstr.cn/32002.14.jfdc.CN10-1649/TP.2025.02.001.

Figures/Tables 13

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Table 1

Fig.7

Table 2

Table 3

Table 4

Table 5

Table 6

References 19

[1]	WANG Z, DONG Y, MAO S, et al. Internet multimedia traffic classification from QoS perspective using semi-supervised dictionary learning models[J]. China Communications, 2017, 14(10): 202-218.
[2]	KALWAR J H, BHATTI S. Deep Learning Approaches for Network Traffic Classification in the Internet of Things (IoT): A Survey[J/OL]. arXiv preprint arXiv:2402.00920, 2024. https://arxiv.org/abs/2402.00920.
[3]	ABBASI M, SHAHRAKI A, TAHERKORDI A. Deep learning for network traffic monitoring and analysis (NTMA): A survey[J/OL]. Computer Communications, 2021, 170: 19-41. https://www.sciencedirect.com/science/article/pii/S0140366421000426.
[4]	YAO H, LIU C, ZHANG P, et al. Identification of encrypted traffic through attention mechanism based long short term memory[J]. IEEE transactions on big data, 2019, 8(1): 241-252.
[5]	WANG X, HAN Y, LEUNG V C M, et al. Edge AI: Convergence of edge computing and artificial intelligence[M]. Singapore: Springer, 2020.
[6]	SIMONYAN K. Very deep convolutional networks for large-scale image recognition[J/OL]. arXiv preprint arXiv:1409.1556, 2014. https://arxiv.org/abs/1409.1556.
[7]	IANDOLA F N. SqueezeNet:AlexNet-level accuracy with 50x fewer parameters and<0.5 MB model size[J/OL]. arXiv preprint arXiv:1602.07360, 2016. https://arxiv.org/abs/1602.07360.
[8]	HOWARD A G. Mobilenets: Efficient convolutional neural networks for mobile vision applications[J/OL]. https://arxiv.org/abs/1602.07360. arXiv preprint arXiv:1704.04861, 2017.
[9]	ZHANG X, ZHOU X, LIN M, et al. Shufflenet: An extremely efficient convolutional neural network for mobile devices[C]// Proceedings of the IEEE conference on computer vision and pattern recognition, 2018: 6848-6856.
[10]	SUN C, CHEN B, BU Y, et al. Lightweight traffic classification model based on deep learning[J]. Wireless Communications and Mobile Computing, 2022, 2022(1): 3539919.
[11]	WANG T, XIE X, WANG W, et al. Netmamba: Efficient network traffic classification via pre-training unidirectional mamba[C]// 2024 IEEE 32nd International Conference on Network Protocols (ICNP), IEEE, 2024: 1-11.
[12]	MA N, ZHANG X, ZHENG H T, et al. Shufflenet v2: Practical guidelines for efficient cnn architecture design[C]// Proceedings of the European conference on computer vision (ECCV), 2018: 116-131.
[13]	R GIL G D, LASHKARI A H, MAMUN M, et al. Characterization of encrypted and VPN traffic using time-related features[C]// Proceedings of the 2nd international conference on information systems security and privacy (ICISSP 2016), Setúbal, Portugal: SciTePress, 2016: 407-414.
[14]	YU F. Multi-scale context aggregation by dilated convolutions[J/OL]. arXiv preprint arXiv:1511.07122, 2015. https://arxiv.org/abs/1511.07122.
[15]	CHEN L C, PAPANDREOU G, KOKKINOS I, et al. Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs[J]. IEEE transactions on pattern analysis and machine intelligence, 2017, 40(4): 834-848.
[16]	HOU Q, ZHOU D, FENG J. Coordinate attention for efficient mobile network design[C]// Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2021: 13713-13722.
[17]	HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]// Proceedings of the IEEE conference on computer vision and pattern recognition, 2016: 770-778.
[18]	HOWARD A, SANDLER M, CHU G, et al. Searching for mobilenetv3[C]// Proceedings of the IEEE/CVF international conference on computer vision, 2019: 1314-1324.
[19]	TAN M, LE Q. Efficientnet: Rethinking model scaling for convolutional neural networks[C]// International conference on machine learning, PMLR, 2019: 6105-6114.

流量类型	流量内容
VPN_Email	Email，Gmail(SMPT，POP3，IMAP)
VPN_Chat	ICQ，AIM，Skype，Facebook，hangouts
VPN_Streaming	Vimeo，Youtube，Netfix，Spotify
VPN_File Transfer	Skype，FTPS，SFTPS
VPN_VoIP	Facebook，Skype，Hangouts，Voipbuster
VPN_P2P	uTorrent，Bittorrent

分辨率	准确率/%	F1值	参数量/M	计算量/M
28×28	96.3	0.961	1.32	3.85
32×32	98.8	0.988	1.32	4.14
36×36	98.2	0.981	1.32	9.15
40×40	98.4	0.982	1.32	9.52

宽度因子	准确率/%	F1值	参数量/M	计算量/M
0.5×	94.3	0.940	0.353	1.17
1×	98.8	0.988	1.32	4.14
1.5×	99.1	0.990	2.91	8.91

模型	准确率/%	F1值	参数量/M	计算量/M
ShuffleNetV2	97.0	0.968	1.26	3.06
ShuffleNetV2+ASPP	97.9	0.978	1.34	8.17
ShuffleNetV2+LM-ASPP	97.3	0.971	1.28	4.05
ShuffleNetV2+RCA	97.9	0.978	1.31	3.45
ShuffleNetV2+LM-ASPP+RCA	98.8	0.988	1.32	4.14

模型	准确率/%	F1值	参数量/M	计算量/M
ResNet-18	98.6	0.985	37.24	11.18
ShuffleNetV2	97.0	0.968	1.26	3.06
ShuffleNetV2+LM-ASPP+RCA	98.8	0.988	1.32	4.14