Title:Enhancing 5G O-RAN Communication Efficiency Through AI-Based Latency Forecasting

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.