Intent-Driven Zero-Touch Provisioning in Private 5G Networks with AI/ML Automation
Shweta Goyal, Krishna Kumar Gattupalli, Bhaksara Rallabandi, Monish Sai Medarametla
Abstract
Industry 4.0, health, logistics, and smart residential campuses are becoming important and have a key infrastructure required in the form of private 5G networks. Nevertheless, their implementation and operation are not simple and have to be modified by experienced human resources to set up, provision, and ensure service-level agreements (SLAs). This paper is a proposal of intent-driven zero-touch provisioning (ZTP) based on AI/ML automation of a private 5G. A mathematical model is created that formalizes intents-to-KPIs mapping and a providing algorithm is presented that is underpinned by simulations of Digital Twins. It is proven through experiment that the proposed framework can cut the provisioning time by 45, decrease the SLA violation by 33, and enhance the automation efficiency in comparison with the baseline rule-based and reactive methods. The findings confirm the practicality of intent-based ZTP using AI/ML as the basis of low-risk and scalable private 5G networks.