Adapting resource utilization according to network needs using MEC, machine learning and network slicing

Abstract

Authors The 5G networks are designed to support a wide range of services that support the variety of requirements, performance and power utilization of services. The rapid increase in mobile network traffic, the number of connected devices, and the variety of applications being implemented require the efficient and dynamic management of network resources. Recently, several technologies have been proposed to improve 5G networks. Indeed, Mobile Edge Computing (MEC) presents a new opportunity to host applications close to users with a reduction in latency and improved performance. In addition, the concept of network slicing facilitates and the construction of multiple logical networks from a single physical infrastructure has been proposed, which optimizes flexible network evolution. In this article, we will seek to optimize the use of network resources, which are scarce resources, to meet the needs of customers and improve the delivery of services on demand by ensuring the smooth operation of services. In this sense, we start by studying the necessary needs in network resources using machine learning technology, then we apply the principle of network slicing based on the requirements of each service to ensure the dynamic and efficient management of network resources. Results show that our method improves resource utilization and ensures flexible, efficient and dynamic sharing of network resources.