Physical Layer Message Prediction for 5G Radio Access Network Protocols

2026-03-20

AI Summary

This work addresses the challenge of automatically reverse-engineering encrypted or unknown communication messages in the 5G physical layer protocol. To this end, it introduces the Transformer architecture—applied for the first time to modeling 5G NR physical layer message sequences. The approach leverages contextual and syntactic structures defined in 3GPP standards and utilizes an open-source 5G system to simulate base station–user equipment interactions, thereby generating labeled training data and constructing a predictive framework. Experimental results demonstrate that the proposed method achieves high-accuracy prediction of subsequent physical layer messages, establishing a novel paradigm for security analysis and reverse engineering of 5G protocols.

Technology Category

Natural Language Processing: GenerationMachine Learning: Large Multimodal Models (LMMs)Planning, Routing, and Scheduling: Planning with Language Models

Application Category

Security and Privacy: Large-scale security measurementsUser Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systemsSearch and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Protocol reverse engineering stands as the cutting-edge approach in security research. This paper presents a framework capable of reverse engineering the communications within a mobile communication system. Our focus is on systems released by the 3GPP, with an emphasis on 5G NR. Our approach leverages the available context and syntax of the 5G standard to predict subsequent messages. This approach relies on a Transformer model and is trained based on an open-source 5G system implementation, emulating a base station and several user equipments. The prediction targets messages at the physical layer.

Problem

• Protocol reverse engineering
• 5G NR
• Physical layer
• Message prediction
• Radio Access Network

Innovation

• Protocol reverse engineering
• 5G NR
• Physical layer prediction
• Transformer model
• 3GPP