City Research Online

Abstract

In the rapidly evolving landscape of digital communications, the demand for fast, reliable, and efficient network traffic management has become more critical than ever. Traditional Internet traffic classification methods, such as payload-based or port-based techniques, are inadequate in the face of encrypted traffic and dynamic port assignments. Addressing these challenges, this thesis presents a comprehensive approach to improving Internet traffic classification by leveraging Artificial Intelligence (AI) with wavelet-transformed trend features, thereby paving the way for more adaptive and accurate solutions in modern network environments.

The core of this research lies in the development of advanced feature extraction techniques using Wavelet Transforms combined with trend analysis. Wavelet Transforms capture both the time and frequency domain characteristics of network traffic, providing a comprehensive view of traffic behavior. By integrating these features with AI models, particularly Machine Learning and Deep Learning algorithms, the thesis enhances the classification accuracy of low-latency traffic, which is critical for applications such as video conferencing, online gaming, and real-time financial transactions. This work contributes to the ongoing efforts to enhance network performance and reliability, ultimately supporting the growing demands of digital applications in today’s interconnected world.

This work also introduces the integration of trend-based features, which reflect the periodicity, seasonality, and long-term behavior of network traffic. These features, when combined with wavelet-transformed data, enhance the robustness of the AI model, making it more resilient to the dynamic nature of Internet traffic. The findings of this research have substantial implications for the design of next-generation network management systems, enabling more efficient allocation of resources and better Quality of Service (QoS) for latency-sensitive applications.

Another considerable contribution of this research is the creation of a dataset and the implementation of an AI algorithms that can dynamically adapt to changing network traffic patterns. Trained on wavelet-transformed trend features, the proposed model captures the unique characteristics of low-latency traffic even in mixed and complex environments. Extensive experiments using real network traffic data to validate the effectiveness of the proposed methods, demonstrating significant improvements in classification accuracy compared to traditional methods.

Publication Type:	Thesis (Doctoral)
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General)
Departments:	School of Science & Technology > Engineering School of Science & Technology > School of Science & Technology Doctoral Theses Doctoral Theses

Export

Downloads