To address the limitations of existing encrypted malicious traffic detection method such as reliance on single-view features and inefficient multi-view feature fusion techniques, the multi-view heterogeneous context graph neural network (MHC-GNN) is proposed for encrypted traffic threat detection. First, in MHC-GNN TLS certificate features and IP features are incorporated into heterogeneous graph modeling alongside traffic statistical features. Second, a multi-path feature fusion mechanism based on meta-paths is employed to dynamically integrate multi-view contextual information and generate highly discriminative flow embedding vectors that enables precise detection of encrypted malicious traffic. Experiments demonstrate that the proposed method achieves an F1 score of 99.7% on CIC-IDS2017 dataset and 95.7% on CIC-DDoS2019 dataset, outperforming multiple baseline graph-based methods overall. This verifies the framework's effectiveness for detecting encrypted malicious traffic.
1)w/o IP Context变体模型出现了最大的性能下滑,F1分数下降了11.6个百分点。这一结果与DDoS攻击的本质高度吻合。CIC-DDoS数据集包含大量分布式拒绝服务攻击,这类攻击特征主要体现为特定源IP或目的IP的高频交互模式。而移除IP节点意味着模型无法聚合IP相关的特征,导致模型在检测DDoS攻击时性能大幅下降。w/o TLS Context变体模型的F1分数下降了5.2个百分点,这说明TLS证书信息能够帮助模型识别特定攻击工具或僵尸网络家族生成的流量。这两种变体证明了仅依靠流的时空特征无法有效识别复杂的攻击模式,引入IP端点和TLS证书的上下文信息是提升检测性能的关键。
ZANGX D, WANGTL, ZHANG,X C,et al. Encrypted malicious traffic detection based on natural language processing and deep learning[J]. Computer Networks, 2024,250:No.110598.
[2]
GUG F, ZHANGJ J, LEEW K. Botsniffer: detecting botnet command and control channels in network traffic[C]//Proceedings of the 2008 Network and Distributed System Security Symposium. San Diego,USA: Internet Society, 2008:269-286.
[3]
The MITRE Corporation. Non-standard port [EB/OL]. (2020-03-14)[2025-10-25].
[4]
SHENM, WEIM W, ZHUL H, et al. Classification of encrypted ttraffic with second-order markov chains and application attribute bigrams[J]. IEEE Transactions on Information Forensics and Security, 2017,12(8):1830-1843.
FUC P, LIQ, SHENM,et al. Realtime robust malicious traffic detection via frequency domain analysis[C]//Proceedings of the 2021 ACM SIGSAC Conference on Computer and Communications Security. New York, USA: ACM, 2021:3431-3446.
SHENM, LIUY T, ZHUL H, et al. Fine-grained webpage fingerprinting using only packet length information of encrypted traffic[J]. IEEE Transactions on Information Forensics and Security, 2021,16:2046-2059.
[10]
SHAOG L, CHENX S, ZENGX M, et al. Deep learning hierarchical representation from heterogeneous flow-level communication data[J]. IEEE Transactions on Information Forensics and Security, 2020,15:1525-1540.
[11]
LINK D, XUX L, GAOH H. TSCRNN: a novel classification scheme of encrypted traffic based on flow spatiotemporal features for efficient management of IIoT[J]. Computer Networks, 2021,190:No.107974.
[12]
SHAPIRAT, SHAVITTY. Flowpic: a generic representation for encrypted traffic classification and applications identification[J]. IEEE Transactions on Network and Service Management, 2021,18(2):1218-1232.
[13]
SHENM, ZHANGJ P, ZHUL H, et al. Accurate decentralized application identification via encrypted traffic analysis using graph neural networks[J]. IEEE Transactions on Information Forensics and Security, 2021,16:2367-2380.
[14]
FUZ Q, LIUM X, QINY, et al. Encrypted malware traffic detection via graph-based network analysis[C]//Proceedings of the 25th International Symposium on Research in Attacks, Intrusions and Defenses. New York, USA: ACM, 2022:495-509.
[15]
HUOHT L, LUOY, LIP L,et al. Flow-based encrypted network traffic classification with graph neural networks[J]. IEEE Transactions on Network and Service Management, 2023,20(2):1224-1237.
[16]
GHADERMAZIJ, HORES, SHAHA,et al. GTAE-IDS: graph transformer-based autoencoder framework for real-time network intrusion detection[J]. IEEE Transactions on Information Forensics and Security, 2025,20:4026-4041.
[17]
WANGW, SHANGY Y, HEY Z, et al. Botmark: automated botnet detection with hybrid analysis of flow-based and graph-based traffic behaviors[J]. Information Sciences, 2020,511:284-296.
[18]
HONGY P, LIQ, YANGY Q,et al. Graph based encrypted malicious traffic detection with hybrid analysis of multi-view features[J]. Information Sciences, 2023,644:No.119229.
[19]
PREMKUMARA, SCHNEIDERM, SPIVEYC,et al. Graph representation learning for context-aware network intrusion detection[C]//Proceedings of SPIE 12538, Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications V. Orlando, USA: SPIE, 2023:No.125380H.
[20]
FUC P, LIQ, XUK. Detecting unknown encrypted malicious traffic in real time via flow interaction graph analysis[DB/OL]. (2023-01-31)[2026-03-03].
[21]
ZHANGH Z, YUL, XIAOX,et al. TFE-GNN: a temporal fusion encoder using graph neural networks for fine-grained encrypted traffic classification[C]//Proceedings of the ACM web conference 2023. New York, USA:ACM, 2023:2066-2075.
[22]
LUOX, YINL H, YANGH Y,et al. SnifferDog: comprehensively learning heterogeneous features of network traffic to identify malicious flows[J]. IEEE Transactions on Information Forensics and Security, 2025,20:11684-11699.
[23]
ZHAOJ, LIQ, HANZ, et al. ReTrial: robust encrypted malicious traffic detection via discriminative relation incorporation and misleading relation correction[J]. IEEE Transactions on Information Forensics and Security, 2025,20:677-692.
[24]
CUIS S, HANX Y, HAND Q, et al. FG-SAT: efficient flow graph for encrypted traffic classification under environment shifts[J]. IEEE Transactions on Information Forensics and Security, 2025,20:5326-5339.
[25]
WANGX, JIH Y, SHIC,et al. Heterogeneous graph attention network[DB/OL]. (2021-01-20)[2026-03-03].
[26]
SHARAFALDINI, HABIBI LASHKARIA, GHORBANIA A. Toward generating a new intrusion detection dataset and intrusion traffic characterization[C]//Proceedings of the 4th International Conference on Information Systems Security and Privacy. Setúbal, Portugal: SCITEPRESS, 2018:108-116.
[27]
SHARAFALDINI, LASHKARIA H, HAKAKS, et al. Developing realistic distributed denial of service (DDoS) attack dataset and taxonomy[C]//Proceedings of the 2019 International Carnahan Conference on Security Technology. Piscataway, USA: IEEE, 2019:1-8.