<\/p>\n\n\n\n
Khalifa University<\/a> of Science and Technology announced the launch of \u2018RF-GPT\u2019, a first-of-its-kind radio-frequency AI language model. The system can interpret wireless signals. It addresses a major limitation in telecommunications AI<\/a>, where language models typically rely only on text and structured network data.<\/p>\n\n\n\n Moreover, RF-GPT showed consistent performance gains in radio-frequency spectrogram tasks. It outperformed baseline models by up to 75.4 percent. This highlights strong radio-frequency understanding. In addition, the model correctly counted signals in a spectrogram nearly 98 percent of the time. This is a capability rarely achieved by general-purpose AI systems.<\/p>\n\n\n\n The model works by converting radio signals into visual patterns. These patterns are then interpreted by artificial intelligence systems. As a result, the system can analyze activity within the wireless spectrum and respond using plain language. This approach supports the UAE National Artificial Intelligence Strategy. It also lays the foundation for more autonomous and intelligent wireless networks.<\/p>\n\n\n\n The project was developed by researchers at Khalifa University. It was led by Merouane Debbah, Senior Director of the Digital Future Institute. The team included postdoctoral fellows Hang Zou and Yu Tian. It also involved research scientists Dr. Lina Bariah, Dr. Samson Lasaulce from Universit\u00e9 de Lorraine, and Dr. Chongwen Huang. PhD student Bohao Wang from Zhejiang University also contributed.<\/p>\n\n\n\n Furthermore, Ahmed Al Durrah, Associate Provost for Research, said the launch reflects a long-term focus on digital infrastructure innovation. He added that it supports AI integration across strategic sectors and next-generation connectivity research. It also aligns with national priorities. He noted that such initiatives help strengthen the UAE\u2019s human capital and research ecosystem.<\/p>\n\n\n\n In addition, Merouane Debbah said RF-GPT marks a turning point for spectrum intelligence. He explained it shifts from isolated, task-specific pipelines to a unified RF-language interface. He added that enabling natural language interaction with the physical layer opens the door to AI-native radio systems. These systems can support network optimization and policy decisions. This is a key step toward future AI-native 6G networks.<\/p>\n\n\n\n RF-GPT was trained on around 625,000 computer-generated radio signal samples. It is designed for telecommunications operators, network engineering teams, and spectrum authorities. The model supports increasingly complex wireless environments. It demonstrated strong performance in identifying signal types and detecting overlapping transmissions. It can also recognize wireless standards and estimate device usage in Wi-Fi networks. Additionally, it can extract data from 5G signals, further reinforcing Khalifa University of Science and Technology\u2019s role in advancing AI-driven telecommunications innovation.<\/p>\n\n\n\n <\/p>\n","protected":false},"excerpt":{"rendered":" Khalifa University of Science and Technology announced the launch of […]<\/p>\n","protected":false},"author":58,"featured_media":105055,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"inline_featured_image":false,"_monsterinsights_skip_tracking":false,"footnotes":""},"categories":[9617],"tags":[73],"contributor":[9732],"class_list":["post-105049","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-latest-news","tag-uae","contributor-news-desk"],"featured_image_src":"https:\/\/techxmedia.com\/en\/wp-content\/uploads\/2026\/04\/RF-GPT_English-copy-42.jpg.jpeg","author_info":{"display_name":"Lubna","author_link":"https:\/\/techxmedia.com\/en\/author\/lubna\/"},"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t