Media Ibrahim
Information System Engineering Department, Erbil Technical Engineering College, Erbil Polytechnic University, Erbil, Iraq

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

Control Traffic in SDN Systems by using Machine Learning techniques: Review Shavan Askar; Diana Hussein; Media Ibrahim; Marwan Aziz Mohammed
International Journal of Research and Applied Technology Vol. 5 No. 1 (2025): Vol 5 No 1 (2025)
Publisher : Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/injuratech.v5i1.15764

Abstract

The rapid development of internet and mobile communication technologies has driven the growth of networking systems, rendering them increasingly complex and diverse in terms of infrastructure, devices, and resources. This complexity necessitates the deployment of network intelligence to enhance system organization, management, maintenance, and optimization. However, applying machine learning (ML) techniques to control and operate networks remains challenging due to the inherently distributed architecture of traditional networks. Conversely, Software-Defined Networking (SDN) facilitates the integration of ML techniques through its centralized operational control, holistic network visibility, software-based traffic monitoring, and dynamic forwarding rule updates. This study presents a comprehensive literature review to address how machine learning techniques can be effectively applied within the SDN framework. First, we outline the foundational background, followed by a concise overview of machine learning methods. We then examine ML applications in SDN, focusing specifically on traffic classification, Quality-of-Service (QoS) prediction, routing optimization, Quality-of-Experience (QoE) management, and resource security. Finally, we discuss current open challenges and future research perspectives.
Deep Learning Security Schemes in IIoT: A Review Shavan Askar; Diana Hussein; Media Ibrahim; Marwan Aziz Mohammed
International Journal of Research and Applied Technology Vol. 5 No. 1 (2025): Vol 5 No 1 (2025)
Publisher : Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/injuratech.v5i1.15851

Abstract

The Industrial Internet of Things (IIoT) is a rapidly expanding technology with the potential to digitize and interconnect diverse sectors, fostering substantial economic opportunities and driving global GDP growth. Driven by the Fourth Industrial Revolution, IIoT platforms generate massive, dynamic, and heterogeneous data streams from interconnected devices and sensors. However, managing data analysis and security across such vast and diverse datasets presents complex challenges. As IIoT adoption expands, cyberattacks become increasingly sophisticated, rendering conventional anomaly detection algorithms less effective. IIoT is widely deployed across manufacturing, logistics, transportation, oil and gas, mining, metallurgy, energy utilities, and aviation. While IIoT offers immense potential for industrial application development, it also heightens vulnerability to cyberattacks and elevates security requirements. The massive volume of data produced by IoT infrastructure demands advanced processing technologies, particularly deep learning (DL). Within IIoT environments, DL algorithms enable smart assembly, smart manufacturing, efficient networking, and proactive accident detection and prevention. These multifaceted applications motivate this article to examine the transformative potential of DL in IIoT systems.