ارایه روش ترکیبی برای شناسایی و طبقه بندی ترافیک در شبکه های بی سیم

نوع مقاله : مقاله پژوهشی

نویسندگان

1 کارشناسی ارشد، دانشکده برق و کامپیوتر، دانشگاه صنعتی مالک اشتر، تهران، ایران

2 استادیار، دانشکده برق و کامپیوتر، دانشگاه صنعتی مالک اشتر، تهران، ایران

چکیده

استفاده از رویکرد اقتضایی با بهره­گیری از ویژگی­هایی از جمله مدیریت توزیع یافته بین گره­ها، تسهیل در امر ورود و خروج آن‌ها به شبکه و امکان تحرک بهتر، یکی از گزینه­های مطلوب جهت پیکربندی شبکه­های بی­سیم می­باشد. همین امر موجب تولید ترافیک با رفتار پوی  توسط نرم­افزارهای کاربردی در چنین شبکه­هایی می­شود که مسئله مدیریت شبکه و کنترل ترافیک بین گره­های را تحت تأثیر خود قرار می­دهد. شناسایی و طبقه­بندی ترافیک جاری در شبکه می­تواند کمک شایانی به این چالش در شبکه­های بی­سیم کند. از آنجا که روش­های مرسوم شناسایی و طبقه­بندی ترافیک قادر به ارائه عملکرد مناسب با چنین ترافیک­هایی نیستند بنابراین استفاده از روش­های مبتنی بر یادگیری ماشین می­تواند برای بهبود طبقه­بندی ترافیک به‌کارگرفته شوند. از آنجا که حساسیت بالا جهت یافتن ترافیک­هایی خاص نیازمند افزایش احتمال آشکارسازی و عدم ارائه تصمیم اشتباه نیازمند کاهش هشدار غلط در سامانه می­باشد،بنابراین در این مقاله روشی جدید جهت افزایش دقت و بهره­وری در شناسایی وطبقه­بندی ترافیک در شبکه­های بی­سیم اقتضایی ارائه می­شود که مبتنی بر ترکیب هدفمند روش­های یادگیری ماشین می­باشد. نتایج نشان می­دهند که روش ارائه شده علاوه بر بهبود معیارهای ارزیابی طبقه­بندی کننده ترافیک موجب افزایش احتمال آشکارسازی و کاهش نرخ هشدار غلط به نسبت به‌کارگیری روش­های یادگیری ماشین به‌صورت یکتا می­باشد.

کلیدواژه‌ها

عنوان مقاله [English]

A New Hybrid Approach for Traffic Identification and Classification in Wireless Networks

نویسندگان [English]

  • Maryam Bazooband 1
  • Hossein Bahramgiri 2
1 Master's degree, Electrical and Computer Faculty, Malek Ashtar University of Technology, Tehran, Iran
2 Assistant Professor, Electrical and Computer Faculty, Malik Ashtar University of Technology, Tehran, Iran
چکیده [English]

Using the ad hoc approach is one of the desirable options for configuration of wireless networks because of the features such as distributed management between nodes, facilitating their entry and exit into the network and the possibility of better mobility. This scheme leads to the dynamic behavior of the traffic generated by applications in such networks, which affects the issue of network management and traffic control between nodes. Identifying and classifying the network traffic can help to deal with these challenges in wireless networks. Because conventional traffic detection and classification methods are not able to provide proper performance with such traffic, applied machine-learning-based methods can improve the detection and classification performance. As the precision required to find a specific network traffic implies a high probability of detection, and the elimination of wrong decisions needs the false alarm rate reduction, in this paper a new hybrid method, based on the combination of machine learning methods is introduced to increase the accuracy and efficiency of identifying and classifying traffic in ad hoc wireless networks based on purposeful combination of various machine learning methods. The results show that in addition to improving the evaluation criteria of traffic classification, the proposed method increases the detection probability and reduces the false alarm rate, in comparison to the cases where only a single machine learning method is used.

کلیدواژه‌ها [English]

  • Wireless Ad hoc Networks
  • Network Traffic
  • Probability of Detection
  • False Alarm Rate
  • Hybrid Approach in Machine Learning

Smiley face

مراجع

[1]    J. Erman, A. Mahanti, and M. Arlitt, "Qrp05-4: Internet Traffic Identification Using Machine Learning," IEEE Globecom,  IEEE, 2006.
[2]    L. T. Hu and LiJun Zhang, "Real-time Internet Traffic Identification Based on Decision Tree," World Automation Congress, IEEE, 2012.
[3]    J. Zhang, C. Chen, Y. Xiang, W. Zhou and A. V. Vasilakos, "An Effective Network Traffic Classification Method with Unknown Flow Detection," IEEE Transactions on Network and Service Management, vol. 10, no.2,  pp. 133-147, 2013.
[4]    A. C. Callado, "A Survey on Internet Traffic Identification," IEEE Communications Surveys and Tutorials, vol. 11, no. 3, pp.  37-52, 2009.
[5]    R. Ma and S. Qin, "Identification of Unknown Protocol Traffic Based on Deep Learning," 3rd IEEE International Conference on Computer and Communications (ICCC), IEEE, 2017.
[6]    Sh. Zuozhi, Y. Yue, and M. Yunlang, "The Research of Protocol Identification Based on Traffic Analysis," 10th International Conference on Intelligent Computation Technology and Automation (ICICTA), IEEE, 2017.
[7]    L. Kong, G. Huang, K. Wu, Q. Tang and S. Ye, "Comparison of Internet Traffic Identification on Machine Learning Methods," International Conference on Big Data and Artificial Intelligence (BDAI), IEEE, 2018.
[8]    S. Kokila, A. Sathish, and R. Shankar, "Comparative Analysis of Internet Traffic Identification Methods," Proceedings of the UGC Sponsored National Conference on Advanced Networking and Applications, 2015.
[9]    L. Peng, "Early Stage Internet Traffic Identification Using Probabilistic Neural Networks," International Journal of Computer and Communication Engineering, vol.  4, no. 6, pp.  417-425, 2015).
[10]  Zh. Wang, "The Applications of Deep Learning on Traffic Identification," BlackHat USA 24, 2015.
[11]  P. Singhal, R. Mathur, and H. Vyas, "Network Traffic Classification Based on Unsupervised Approach," International Journal of Computer Applications, 975, 8887, 2013.
[12]  R. Thupae, B. Isong, N. Gasela and A. M. Abu-Mahfouz, "Machine Learning Techniques for Traffic Identification and Classifiacation in SDWSN: A Survey," IECON 2018-44th Annual Conference of the IEEE Industrial Electronics Society, IEEE, 2018.
[13]  Y. Liu, W. Li, and Y. Ch. Li, "Network Traffic Classification Using K-means Clustering," Second International Multi-Symposiums on Computer and Computational Sciences (IMSCCS 2007), IEEE, 2007.
[14]  Ch. GU and Sh. Zhang, "Online Wireless Mesh Network Traffic Classification Using Machine Learning,” Journal of Computational Information Systems, 2011.
[15]  J. Ran and Y. Chen, "Three Dimensional Convolutional Neural Network Based Traffic Classification for Wireless Communications," pp. 624-627, [M1] 2018,
[16]  W. Wang, M. Zhu, X. Zeng, X. Ye, and Y. Sheng, "Malware Traffic Classification Using Convolutional Neural Network for Representation Learning," ICOIN, IEEE, 2017.
[17]  F. Noorbehbahani and S. Mansoori, "A New Semi-Supervised Method for Network Traffic Classification Based on X-Means Clustering and Label Propagation," 8th International Conference on Computer and Knowledge Engineering (ICCKE), Mashhad, pp. 120-125, 2018.
[18] C. M. Bishop,  “Pattern Recognition and Machine Learning (Information Science and Statistics),” Springer, 2007. ISBN: 0387310738https://github.com/echowei/DeepTraffic/tree/master/1.malware_traffic_classif    ication/1.DataSet(USTC-TFC2016)/Benigen. 
پدافند الکترونیکی و سایبری
دوره 10، شماره 2 - شماره پیاپی 38
شماره پیاپی 38، فصلنامه تابستان
مهر 1401
صفحه 31-41

فایل ها

سابقه مقاله

  • تاریخ دریافت: 30 فروردین 1400
  • تاریخ بازنگری: 20 شهریور 1400
  • تاریخ پذیرش: 07 مهر 1400
  • تاریخ انتشار: 01 مهر 1401

هم رسانی

ارجاع به این مقاله

آمار