REAL-TIME NETWORK TRAFFIC ANALYSIS USING RECURRENT NEURAL MODELS: REAL-TIME NETWORK TRAFFIC ANALYSIS USING RECURRENT NEURAL MODELS

Daniel Denev; Liliya  Staneva

doi:10.46687/jsar.v28i1.439

Authors

Daniel Denev DEPARTMENT OF COMMUNICATION AND COMPUTER ENGINEERING AND SECURITY TECHNOLOGIES, FACULTY OF TECHNICAL SCIENCES, KONSTANTIN PRESLAVSKY UNIVERSITY OF SHUMEN, SHUMEN 9712, 115, UNIVERSITETSKA STR., E-MAIL:d.denev@shu.bg
Liliya Staneva BURGAS STATE UNIVERSITY "PROF. DR. ASSEN ZLATAROV", E-MAIL: anestieva@mail.bg

DOI:

https://doi.org/10.46687/jsar.v28i1.439

Keywords:

Neural models, Network traffic, Computer networks, Communication, Monitoring

Abstract

With the increasing complexity and dynamics of modern computer networks, there is a need for methods that can provide timely and accurate analysis of network traffic. Traditional techniques for monitoring and anomaly detection are often limited by static rules and the inability to capture complex time dependencies. In this context, recurrent neural networks (RNNs), and in particular their advanced architectures such as LSTM and GRU, appear as promising tools for real-time analysis. This article reviews the application of recurrent neural models for prediction, classification and anomaly detection in network traffic, focusing on architectural features, challenges and potential directions for development.

Author Biographies

Daniel Denev, DEPARTMENT OF COMMUNICATION AND COMPUTER ENGINEERING AND SECURITY TECHNOLOGIES, FACULTY OF TECHNICAL SCIENCES, KONSTANTIN PRESLAVSKY UNIVERSITY OF SHUMEN, SHUMEN 9712, 115, UNIVERSITETSKA STR., E-MAIL:d.denev@shu.bg

DEPARTMENT OF COMMUNICATION AND COMPUTER ENGINEERING AND SECURITY TECHNOLOGIES, FACULTY OF TECHNICAL SCIENCES, KONSTANTIN PRESLAVSKY UNIVERSITY OF SHUMEN, SHUMEN 9712, 115, UNIVERSITETSKA STR., E-MAIL:d.denev@shu.bg

Liliya Staneva, BURGAS STATE UNIVERSITY "PROF. DR. ASSEN ZLATAROV", E-MAIL: anestieva@mail.bg

BURGAS STATE UNIVERSITY "PROF. DR. ASSEN ZLATAROV", E-MAIL: anestieva@mail.bg

References

Bakhshi T., Anomaly Detection in Encrypted Internet Traffic Using Hybrid Deep Learning, 2021, Security and Communication Networks, Hindawi, DOI: 10.1155/2021/5363750.

Guo S., Liu Y., Su Y., Network Traffic Anomaly Detection Method Based on CAE and LSTM, 2021, Journal of Physics: Conference Series, IOP Publishing, DOI: 10.1088/1742-6596/2025/1/012025.

Nguyen D., Real-Time Threat Detection Using Network Flow Analysis and LSTM Networks, 2020, International Journal of Information Technology and Computer Engineering, Vol. 8, No. 4, DOI: 10.62647/.

Riaz H., Hussain Z., Hasan Z., Mustafa M., Federated Learning for Distributed Anomaly Detection in Network Traffic Using GRU-Based Models, 2025, Spectrum of Engineering Sciences, Vol. 3, No. 3, pp. 522–534.

Sharma A., Singh R., Kaur J., Improved Network Anomaly Detection System Using Optimized Autoencoder–LSTM, 2025, Expert Systems with Applications, Vol. 273, DOI: 10.1016/j.eswa.2025.126854.

REAL-TIME NETWORK TRAFFIC ANALYSIS USING RECURRENT NEURAL MODELS