Dynamic Network Slice Orchestration using Gated Recurrent Units and Transformer-based Hybrid Model for IoT

Dynamic network slicing has become a cornerstone in the evolution of modern telecommunications and networking. In an increasingly diverse and data-intensive digital landscape, the ability to adapt network resources in real-time has never been more crucial. This paper explores the significance of dynamic slicing in meeting the evolving demands of telecommunications, particularly in the context of emerging technologies such as 5G, Internet of Things (IoT), and edge computing. This paper introduces a novel hybrid GRU-Transformer dynamic slicing mechanism designed to optimize resource allocation, minimize packet loss, and ensure reliable network performance across a wide range of scenarios. By combining the strengths of a GRU and a Transformer, this mechanism offers adaptability, real-time monitoring, and predictive analytics, making it a valuable tool for enhancing network efficiency and responsiveness in the face of dynamic demands. The proposed model achieves the highest service utilization of 98.2%, demonstrating the capabilities of the hybrid GRU-Transformer mechanism in addressing the challenges of modern network management.