Enhancing 5G O-RAN Communication Efficiency Through AI-Based Latency Forecasting
Raúl Parada, Ebrahim Abu-Helalah, Jordi Serra, Anton Aguilar, Paolo Dini
TL;DR
The paper tackles latency management in dynamic 5G O-RAN by integrating an AI-driven latency forecasting system into a functional, open-source prototype built on FlexRIC. It deploys a bidirectional LSTM latency predictor that operates in real time within a containerized O-RAN stack, validated on hardware with a 56k-sample dataset achieving low prediction loss. The results show accurate latency forecasting that informs transmission decisions, improving resource efficiency and reliability in a scalable, replicable deployment. This work bridges the gap between ML-based latency forecasting and real-world hardware validation in O-RAN, offering a practical path toward smarter, delay-sensitive 5G networks.
Abstract
The increasing complexity and dynamic nature of 5G open radio access networks (O-RAN) pose significant challenges to maintaining low latency, high throughput, and resource efficiency. While existing methods leverage machine learning for latency prediction and resource management, they often lack real-world scalability and hardware validation. This paper addresses these limitations by presenting an artificial intelligence-driven latency forecasting system integrated into a functional O-RAN prototype. The system uses a bidirectional long short-term memory model to predict latency in real time within a scalable, open-source framework built with FlexRIC. Experimental results demonstrate the model's efficacy, achieving a loss metric below 0.04, thus validating its applicability in dynamic 5G environments.