Abu Dhabi’s Khalifa University of Science and Technology has announced the launch of ‘RF-GPT’, a new radio-frequency AI language model developed by its Digital Future Institute. The model is capable of interpreting wireless signals, a significant step forward for telecom AI which has historically been limited to text and structured network data.
Quick Facts
- Analyzes radio signals into understandable language.
- Outperforms existing models by up to 75.4%.
- Correctly counts signals with ~98% accuracy.
How RF-GPT Works
RF-GPT functions by converting raw radio signals into visual patterns that artificial intelligence can process. Once translated into this visual format, the system can analyze the patterns and answer questions in plain language about activity within the wireless spectrum.
The model has shown strong performance in radio frequency understanding, particularly in spectrogram tasks. For instance, RF-GPT correctly counted the number of signals present in a spectrogram approximately 98% of the time, a task where general-purpose AI models typically fail. This development directly contributes to the UAE National Artificial Intelligence Strategy by building a foundation for more autonomous and intelligent wireless networks.
A New Approach to Network Intelligence
The project reflects a shift in how AI can be applied to physical infrastructure, creating new possibilities for network management and optimization.
Professor Merouane Debbah, Senior Director of the Digital Future Institute, explained the model’s impact: “RF-GPT represents a turning point for spectrum intelligence, moving from isolated, task-specific radio frequency pipelines toward a unified RF-language interface. We gave a language model its first glimpse of the electromagnetic spectrum and the view is already remarkable. By making the physical layer quarriable in natural language, we open the door to AI-native radio systems where RF perception can directly support network optimization and policy decisions, a crucial step toward future AI-native 6G networks.”
Professor Ahmed Al Durrah, Associate Provost for Research at Khalifa University, added, “The launch of ‘RF-GPT’ reflects Khalifa University’s long-term focus on innovation in digital infrastructure to advance AI integration across strategic sectors, and next-generation connectivity research, aligned with national priorities.”
Training and Applications
RF-GPT was trained on a dataset of approximately 625,000 computer-generated radio signal examples. It is designed for use by telecom operators, network engineering teams, and spectrum authorities who manage increasingly complex wireless environments.
The model has demonstrated strong performance across several key tasks, including identifying different signal types, detecting overlapping transmissions, recognizing wireless standards, estimating device usage on Wi-Fi networks, and extracting data from 5G signals.
About the Digital Future Institute
The Digital Future Institute (DFI) is Khalifa University’s applied AI and ICT institute dedicated to designing, building, and deploying intelligent digital systems across communications, networks, energy, climate, and secure infrastructure. The Institute combines foundational research with industry partnerships to accelerate the development of foundation models and deployable AI platforms. DFI aims to bridge academia and industry to translate AI research into operational, real-world systems.
Source: Zawya
