Darknet tabanlı CNN'ler ile sinyal seviyesinde ağ trafiği sınıflandırması: Yeni bir yöntemsel yaklaşım

Yıl 2026, Cilt: 41 Sayı: 1 , 383 - 400 , 31.03.2026

Munip Geylani , Musa Çıbuk , Ayhan Akbal

https://doi.org/10.17341/gazimmfd.1761166

https://izlik.org/JA64FN36LP

Öz

Ağ trafiği sınıflandırması, araştırmacılar tarafından yoğun biçimde çalışılan önemli bir alandır. Mevcut çalışmaların genellikle veri bağı katmanı ve üzerindeki katmanlarda, trafiğin çözümlenmiş (decode edilmiş) hali üzerinde yapıldığı görülmektedir. Bu çalışmada ise ağ trafiği sınıflandırmasına fiziksel katman düzeyinde, yeni bir sınıflandırma yaklaşımı sunulmaktadır. Önerilen yöntemde, ağ trafiği elektriksel sinyaller aracılığıyla temsil edilmekte ve sınıflandırma işlemi bu ham sinyaller üzerinden gerçekleştirilmektedir. Bu amaçla, bilgisayar ile ağ anahtarlama cihazı arasındaki fiziksel bağlantıdan, osiloskop yardımıyla sinyaller toplanmış ve etiketlenerek 6 farklı trafik türünü içeren yeni bir veri seti oluşturulmuştur. Oluşturulan sinyal veri seti, yatay, spiral, diyagonal zikzak ve spektrogram teknikleriyle görselleştirilerek Darknet tabanlı CNN mimarilerinin eğitilmesinde kullanılmıştır. Darknet53 ve diyagonal zikzak görselleştirme tekniğiyle yapılan sınıflandırma deneyinde %96,24 doğruluk oranı ile en yüksek performans elde edilmiştir. Elde edilen sonuçlar, paket içerikleri çözülmeden ve veri mahremiyeti ihlal edilmeden, ağ trafiğinin sinyal düzeyinde dahi yüksek doğrulukla sınıflandırılabileceğini göstermektedir. Bu yönüyle çalışma, geleneksel yöntemlere güçlü bir alternatif sunmaktadır.

Anahtar Kelimeler

Ağ trafiği sınıflandırması , sinyal seviyesinde ağ trafiği sınıflandırma , fiziksel katman , sinyal görselleştirme , Darknet

Signal-level network traffic classification using darknet-based CNNs: A new methodological approach

https://izlik.org/JA64FN36LP

Öz

Network traffic classification is a significant research area that has been extensively studied by researchers. Existing studies are mostly conducted at the data link layer or higher layers, typically using the decoded form of the traffic. In this study, a novel classification approach is proposed at the physical layer level. In the proposed method, network traffic is represented through electrical signals, and the classification process is performed directly on these raw signals. To this end, signals were collected from the physical connection between a computer and a network switch using an oscilloscope, and a new dataset consisting of six different traffic types was constructed by labeling these signals. The created signal dataset was visualized using horizontal, spiral, diagonal zigzag, and spectrogram techniques, and was then used to train Darknet-based CNN architectures. The highest performance was achieved in the classification experiment using Darknet53 and the diagonal zigzag visualization technique, with an accuracy rate of 96.24%.The results demonstrate that network traffic can be effectively classified at the signal level, without decoding packet contents and without compromising data privacy. In this respect, the study offers a strong alternative to traditional classification methods.

Anahtar Kelimeler

Network traffic classification , signal-level traffic classification , physical layer , signal visualization , Darknet

Kaynakça

• 1. Wang Y., Yun X., Zhang Y., Zhao C., Liu X, A Multi-Scale Feature Attention Approach to Network Traffic Classification and Its Model Explanation, IEEE Trans Netw Serv Manag, 19 (2), 875–889, 2022.
• 2. Rezaei S., Liu X., Deep Learning for Encrypted Traffic Classification: An Overview, IEEE Commun Mag, 57 (5), 76–81, 2019.
• 3. Abbasi M., Shahraki A., Taherkordi A., Deep Learning for Network Traffic Monitoring and Analysis (NTMA): A Survey, Comput Commun, 170(February), 19–41, 2021.
• 4. Zheng Y., Dang Z., Lian X., Peng C., Gao X., Multi-view multi-label network traffic classification based on MLP-Mixer neural network, Comput Networks, 253(August), 110746, 2024.
• 5. Chiu K.C., Liu C.C., Chou L. Der., CAPC: Packet-Based Network Service Classifier with Convolutional Autoencoder, IEEE Access, 8, 218081–218094, 2020.
• 6. Zhao J., Jing X., Yan Z., Pedrycz W., Network traffic classification for data fusion: A survey, Inf Fusion, 72, 22–47, 2021.
• 7. Azab A., Khasawneh M., Alrabaee S., Choo K.K.R., Sarsour M., Network traffic classification: Techniques, datasets, and challenges, Digit Commun Networks, 10 (3), 676–692, 2024.
• 8. Ahmed A.A., Agunsoye G., A real-time network traffic classifier for online applications using machine learning, Algorithms, 14 (8), 250, 2021.
• 9. Pacheco F., Exposito E., Gineste M., Baudoin C., Aguilar J., Towards the Deployment of Machine Learning Solutions in Network Traffic Classification: A Systematic Survey, IEEE Commun Surv Tutorials, 21 (2), 1988–2014, 2019.
• 10. Burukanlı M., Çıbuk M., Intrusion Detection and Performance Analysis Using Copula Functions, Bitlis Eren Üniversitesi Fen Bilim Derg, 13 (4), 1335–1354, 2024.
• 11. Zhou D., Yan Z., Fu Y., Yao Z., A survey on network data collection, J Netw Comput Appl, 116(May), 9–23, 2018.
• 12. Hu Y., Zeng Z., Song J., Xu L., Zhou X., Online network traffic classification based on external attention and convolution by IP packet header, Comput Networks, 252(June), 110656, 2024.
• 13. Fang Z., Gao X., Zhang H., Tang J., Gao Q., Application Layer Protocol Identification Method Based on ResNet, Algorithms, 18 (1), 52, 2025.
• 14. Lotfollahi M., Jafari S.M., Shirali H.Z.R., Saberian M., Deep packet: a novel approach for encrypted traffic classification using deep learning, Soft Comput, 24 (3), 1999–2012, 2020.
• 15. Zhou H., Wang Y., Lei X., Liu Y., A Method of Improved CNN Traffic Classification, In: 2017 13th Int. Conf. Comput. Intell. Secur. IEEE, 177–181, 2017.
• 16. Yamansavascilar B., Guvensan M.A., Yavuz A.G., Karsligil M.E., Application identification via network traffic classification, In: 2017 Int. Conf. Comput. Netw. Commun. IEEE, 843–848, 2017.
• 17. Salman O., Elhajj I.H., Chehab A., Kayssi A., A Multi-level Internet Traffic Classifier Using Deep Learning, Proc 2018 9th Int Conf Netw Futur NOF 2018, 68–75, 2018.
• 18. Lopez-Martin M., Carro B., Sanchez-Esguevillas A., Lloret J., Network Traffic Classifier With Convolutional and Recurrent Neural Networks for Internet of Things, IEEE Access, 5, 18042–18050, 2017.
• 19. Al-Jameel M., Turner S., Kanakis T., Al-Sherbaz A., Bhaya W.S., Deep Learning Approach for Real-time Video Streaming Traffic Classification, In: 2022 Int. Conf. Comput. Sci. Softw. Eng. IEEE, 168–174, 2022.
• 20. Aouini Z., Kortebi A., Ghamri-Doudane Y., Cherif I.L., Early classification of residential networks traffic using C5.0 machine learning algorithm, In: 2018 Wirel. Days. IEEE, 46–53, 2018.
• 21. Fauvel K., Chen F., Rossi D., A Lightweight, Efficient and Explainable-by-Design Convolutional Neural Network for Internet Traffic Classification, In: Proc. 29th ACM SIGKDD Conf. Knowl. Discov. Data Min. ACM, New York, NY, USA, 4013–4023, 2023.
• 22. Wang Y., Gao Y., Li X., Yuan J., Encrypted Traffic Classification Model Based on SwinT-CNN. 2023 4th Int Conf Comput Eng Appl ICCEA 2023, 138–142, 2023.
• 23. Zheng Y., Dang Z., Lian X., Peng C., Gao X., Multi-view multi-label network traffic classification based on MLP-Mixer neural network. Comput Networks, 253(August), 110746, 2024.
• 24. Yang Y., Yan Y., Gao Z., Rui L., Lyu R., Gao B., Yu P., A Network Traffic Classification Method Based on Dual-Mode Feature Extraction and Hybrid Neural Networks. IEEE Trans Netw Serv Manag, 20 (4), 4073–4084, 2023.
• 25. Kırışoğlu S., Kotan B., Kotan K., Çok Katmanlı Algılayıcı ile Ağ Trafiği Sınıflandırma Analizi, Düzce Üniversitesi Bilim ve Teknol Derg, 10 (2), 837–846, 2022.
• 26. Shams M., Ağ anomali̇si̇ tespi̇ti̇nde maki̇ne öğrenmesi̇ algori̇tmalarının kullanımı ve karşılaştırmalı anali̇zi̇, Yüksek Lisans Tezi, Sakarya Üniversitesi, Fen Bilimleri Enstitüsü, Sakarya, 2020.
• 27. Özekes S., Karakoç E.N., Makine Öğrenmesi Yöntemleriyle Anormal Ağ Trafiğinin Tespit Edilmesi, Düzce Üniversitesi Bilim ve Teknol Derg, 7 (1), 566–576, 2019.
• 28. Khan A.S., Ahmad Z., Abdullah J., Ahmad F., A Spectrogram Image-Based Network Anomaly Detection System Using Deep Convolutional Neural Network, IEEE Access, 9, 87079–87093, 2021.
• 29. Salati M., Askerzade İ., Bostancı G.E., Convolutional neural network models using metaheuristic based feature selection method for intrusion detection, Journal of the Faculty of Engineering and Architecture of Gazi University, 40 (1), 179–188, 2024.
• 30. Wang W., Sheng Y., Wang J., Zeng X., Ye X., Huang Y., Zhu M., HAST-IDS: Learning Hierarchical Spatial-Temporal Features Using Deep Neural Networks to Improve Intrusion Detection, IEEE Access, 6, 1792–1806, 2018.
• 31. Imrana Y., Xiang Y., Ali L., Abdul-Rauf Z., A bidirectional LSTM deep learning approach for intrusion detection, Expert Syst Appl, 185 (June), 115524, 2021.
• 32. Yu L., Dong J., Chen L., Li M., Xu B., Li Z., Qiao L., Liu L., Zhao B., Zhang C., PBCNN: Packet Bytes-based Convolutional Neural Network for Network Intrusion Detection, Comput Networks, 194 (January), 108117, 2021.
• 33. Hu X., Gu C., Chen Y., Wei F., tCLD-Net: A Transfer Learning Internet Encrypted Traffic Classification Scheme Based on Convolution Neural Network and Long Short-Term Memory Network, In: 2021 Int. Conf. Commun. Comput. Cybersecurity, Informatics, IEEE, 1–5, 2021.
• 34. Uǧurlu M., Doǧru I.A., Arslan R.S., Detection and classification of darknet traffic using machine learning methods, Journal of the Faculty of Engineering and Architecture of Gazi University, 38 (3), 1737–1746, 2023.
• 35. Wang W., Zhu M., Wang J., Zeng X., Yang Z., End-to-end encrypted traffic classification with one-dimensional convolution neural networks, In: 2017 IEEE Int. Conf. Intell. Secur. Informatics, IEEE, 43–48, 2017.
• 36. Bit-Twist: a flexible packet generator and editör, https://bittwist.sourceforge.io/doc.html, Erişim tarihi Ekim 11, 2024
• 37. Geylani M., Çıbuk M., Akbal A., Ethernet Frame Physical-Layer Signal Dataset - 10BaseT, Mendeley Data, https://data.mendeley.com/datasets/x8x39r6nmt/1, 2025.
• 38. Demir Ş.N., Çıbuk M., The impact of signal visualization types on the performance of image processing-based convolutional neural networks, In: Akdeniz 14th Int. Conf. Appl. Sci., Academy Global Publishing House, Kyrenia, 98–126, 2025.
• 39. Şeker A., Diri B., Balık H., A review about deep learning methods and applications, Gazi Journal of Engineering Sciences, 3 (3), 47-64, 2017.
• 40. Amri A.A., Ismail A.R., Zarir A.A., Comparative performance of deep learning and machine learning algorithms on imbalanced handwritten data, Int J Adv Comput Sci Appl, 9 (2), 258–264, 2018.
• 41. Redmon J., Farhadi A., YOLO9000: Better, Faster, Stronger, In: 2017 IEEE Conf. Comput. Vis. Pattern Recognit, IEEE, 6517–6525, 2017.
• 42. Redmon J., Farhadi A., YOLOv3: An Incremental Improvement, arXiv Prepr arXiv180402767, 1–6, 2018.
• 43. Wu W., Guo L., Gao H., You Z., Liu Y., Chen Z., YOLO-SLAM: A semantic SLAM system towards dynamic environment with geometric constraint, Neural Comput Appl, 34 (8), 6011–6026, 2022.