Muhammad Fendi Osman, Mohd Rizal Bin Mohd Isa, Mohammad Adib Khairuddin, Mohd Afizi Mohd Shukran, Noor Afiza Mat Razali, Nur Diyana Binti Kamarudin, Kamaruzaman Maskat, Roziyani Rawi, Hendra Hidayat
Dynamic Quality of Service (QoS) in computer networks refers to techniques, mechanisms, or technologies that have the ability to automatically adjust QoS requirements or parameters to optimize network performance in real-time without the need for human intervention and manual configuration. As networks grow, maintaining optimal QoS capabilities in the face of high, fluctuating traffic loads and varying application demands becomes increasingly important and challenging. Based on the dynamic QoS that has been developed by the author through previous publications, this paper introduces a new enhancement that leverages deep learning (DL) techniques to analyse bandwidth and jitter (input parameters) to optimise the network automatically, more sophisticatedly, and responsively. Unlike traditional methods that rely on static rules or simple conditional logic, our proposed solution uses a neural network model to predict bandwidth allocation requirements based on real-time network conditions. Indirectly, the proposed solution is significantly able to improve the adaptability and efficiency of the QoS adjustment. Through comprehensive simulation and testing in a Software Defined Network (SDN) test environment that has been developed using the MiniEdit (Mininet Graphical Emulator), the results show that our DL-Enhanced Dynamic QoS Framework not only outperforms conventional QoS management approaches in terms of latency and throughput but also ensures a higher quality of experience for end users. This research highlights the potential for integrating artificial intelligence (AI) with network management practices to address the complexities of modern network infrastructure and pave the way for more resilient and user-centric network solutions. © School of Engineering, Taylor’s University.
National Defence University of Malaysia, Faculty of Defence Science and Technology, Level 2 Bangunan Lestari, Kem Perdana Sungai Besi, Kuala Lumpur, 57000, Malaysia; Department of Electronics Engineering, Universitas Negeri Padang, Padang, Indonesia
