The Development of Framework Design Using AI for VC in 5G N/W

This paper provides an introduction to the 5G mobile network testbed, with the proposed idea of using containers for architectures’ virtualization and orchestration. This testbed is integrated using open technologies and methods for the flexible deployment of virtual network functions under the scenario of applying integrated artificial intelligence features to boost the performance of an AI-driven network and, hence, reduce costs. For example, this work presents the versatile testbed architecture engineered to support the integration of AI agents for performance enhancement of mobile network functionalities. The performance of the testbed and its relevance are validated by the performance evaluation of two use cases, namely RAN slicing and VNF placement. For example, RAN-slicing applications take advantage of FlexRAN and Ryu controllers to dynamically allocate resources based on slice requirements, whereas supervised learning ensures an optimal resource allocation. This study has proved that AI-driven resource allocation ensures throughput and SLA requirements much better than traditional methods. Similarly, the testbed supports VNF placement applications, where several placement scenarios are evaluated and the VNF placement is dynamically altered in accordance with the output of ML models to offer massive improvement in latency and meet SLA requirements. The overall results of the evaluation are a demonstration of the potential of the testbed to simulate and test AI-driven enhancements in mobile networks by providing a flexible environment to research on many AI algorithms and methodologies in optimizing resource allocation and enhancing network performance.