Abstract The fast expansion of Internet of Things (IoT) devices an‎d the global implementation of fifth generation (5G) networks make hyper-connected systems possible in ways that have never been seen before. But this convergence creates big problems for traffic engineering, especially when it comes to meeting the different Quality of Service (QoS) needs of massive Machine-Type Communications (mMTC) an‎d Ultra-Reliable Low-Latency Communications (URLLC). To properly manage these different traffic patterns, traditional distributed networking architectures donʹt have the centralized intelligence an‎d programmability they need. Software-Defined Networking (SDN) is a well-known solution, however most current models donʹt give clear instructions on how to han‎dle QoS at the same time, which wastes resources. This paper looks at how SDN can improve traffic engineering in 5G-IoT networks to fill this vacuum in research. This study uses a systematic literature review an‎d a comparison of the best current research to look at modern SDN-based technologies an‎d measure how much they enhance important QoS measures including latency, jitter, throughput, an‎d packet loss. It shows how SDNʹs ability to control an‎d program resources from one place makes it easier to manage resources intelligently for a wide range of IoT tasks. The study reveals that SDN based dynamic control is essential for successfully han‎dling mMTC traffic an‎d achieving the strict latency requirements of URLLC apps (less than 1ms). In the end, the results show that SDN is a basic architecture for making networks that are smarter, more scalable, an‎d conscious of quality of service (QoS). This opens the door for future improvements that use technologies like machine learning an‎d blockchain to make network management completely automatic.

10/29/2025 12:00:00 AM

87998

1404/08/08

Traffic Engineering for IoT Networks Using SDN in 5G Architecture

IoT5G , Software-Defined Networking (SDN) , Traffic Engineering , Quality of Service (QoS) , Network Slicing , Edge Computing