DeepSig experts in artificial intelligence (AI) and machine learning (ML) for wireless communications announced it completed the world’s first AI-Native 5G over-the-air call. Featuring a commercial 5G handset and 5G standalone network to data services on the public internet, the call is a major industry milestone that proves the benefits of AI-Native wireless systems for 4G, 5G and 6G. The call was achieved in DeepSig’s Wireless AI Innovation Lab.
AI and ML enable 5G networks to be predictive and proactive, including intelligent base stations that make decisions for themselves. This will improve the efficiency, latency, and reliability of network applications. For example, ML within the RAN minimizes power consumption, leverages channel information to support improved connections and enables full use of Massive MIMO systems.
Recommended News : Predictions Series 2022: Interview with Param Kahlon, Chief Product Officer at UiPath
DeepSig’s patented AI ML software significantly enhances 5G baseband performance and at lower costs compared to traditional methods used in mobile networks today. For the call demonstration, DeepSig replaced specific radio receiver algorithms within a 5G vRAN distributed unit (DU) with an AI neural network trained “natively” from actual radio environments. This completes over-the-air calls from commercial user equipment (UE) through the radio unit (RU) connected to the AI-enhanced DU, a standard central unit (CU) and 5G core.
“This industry-first AI-Native 5G over-the-air call is a major milestone in using AI and ML to transform mobile networks and the user experiences they provide,” said James Shea, DeepSig CEO. “AI-Native 5G enables networks to work smarter and continually improve, which means they use less energy and provide faster, more reliable connections. DeepSig’s pioneering work on AI-Native 5G also sets the stage for 6G standards to make AI and ML a fundamental part of tomorrow’s networks.”
Recommended AI News: C Ventures Backed RTFKT to be Acquired by NIKE
[To share your insights with us, please write to sghosh@martechseries.com]
Comments are closed.