LLM-enabled intent-driven service configuration for next generation networks

Intent-Based Networking (IBN) is a promising paradigm for next generation networks, enabling automated network management based on user-defined business network

requirements (Intents). However, current IBN approaches consider that users require expertise in some formal and technical models (e.g., Network Service Descriptors - NSDs) to define these Intents, necessitating substantial effort. A natural progression

of IBN systems is to define Intents using natural language instead of structured models. However, dealing with this becomes challenging due to the unstructured and ambiguous nature of natural language. Fortunately, Large Language Models (LLMs) are becoming very powerful in understanding human language, making them well-suited for this task. This paper proposes an LLM-based Intent translation system that allows users to express

Intents in natural language, which the system subsequently converts into NSDs. Moreover, we employ a Human Feedback (HF) loop that enables the system to learn from past experiences. Evaluations conducted at the EURECOM 5G facility [1] confirm

the effectiveness of our approach in generating accurate NSDs suitable for deployment on an edge computing cluster.