Deutsche Telekom, Nokia and NVIDIA are validating how AI-RAN can improve radio performance and deliver a better mobile experience for customers. Initial lab measurements of AI-based uplink channel estimation demonstrated typical throughput improvements of 10–15%, with gains increasing as radio channel complexity increased.
The validation combines Nokia commercial 5G RAN software with NVIDIA Aerial RAN Computer and applies an AI model to perform uplink channel estimation in real time alongside the radio workload.
The results demonstrate the potential to introduce AI directly into the 5G radio receiver signal processing chain to improve uplink throughput for the best customer experience. Following successful lab validation, the companies plan to validate the technology under real-world conditions in a field trial.
“At Deutsche Telekom, we focus on AI that delivers tangible benefits for our customers. The promising lab results with Nokia and NVIDIA on AI-based channel estimation show the potential to improve network performance and efficiency. Next, we plan to validate these gains in a field trial,” said Arash Ashouriha, SVP Group Technology, Deutsche Telekom. “Together with Nokia and NVIDIA, we will build on this work in our 6G Innovation Hub, exploring further AI-driven capabilities to make mobile networks more intelligent, adaptive and efficient.”
“AI-RAN is moving from research to measurable operator value. Achieving typical uplink throughput gains of 10–15% with Nokia commercial 5G RAN software demonstrates how AI can help operators get more performance from the spectrum and networks they already own. Together with Deutsche Telekom, we are now taking the next step towards field validation,” commented Emma Falck, President of Mobile Infrastructure at Nokia.
“Deutsche Telekom is pioneering a full-stack distributed infrastructure that brings intelligence from the cloud to the network edge,” said Ronnie Vasishta, Senior Vice President of Telecom, NVIDIA. “AI-RAN is a key part of that vision. Together with Nokia, we’re demonstrating at Deutsche Telekom how NVIDIA accelerated computing can deliver measurable gains in radio performance in the lab today, while laying the foundation for AI-native 6G networks.”
The initial work focuses on AI-based uplink channel estimation for the Physical Uplink Shared Channel (PUSCH); which carries uplink user data over the air interface.
As radio signals travel between a user device and the network, buildings, vehicles and other obstacles can reflect, block and scatter them. Channel estimation enables the RAN to account for these changing radio conditions and accurately process the received signal.
Instead of relying solely on conventional signal-processing algorithms, the solution uses a real-time AI model to estimate the radio channel. The AI model runs on GPU-based compute alongside the RAN workload, demonstrating how AI can become an integral part of real-time radio processing.
Measurements conducted in Deutsche Telekom’s lab in Bonn and Nokia’s lab in Oulu showed uplink throughput improvements of 10–15% with the tested AI-based channel estimation functionality. The gains increased under more complex radio channel conditions, indicating the potential of AI to enhance performance particularly in challenging network environments.
The work forms part of Deutsche Telekom and Nokia’s broader exploration of AI-for-RAN: applying AI to optimize network performance, efficiency and ultimately customer experience.
The next phase will move from controlled lab environments towards a field trial on commercial systems to validate how AI-based channel estimation performs under real-world radio conditions.
Deutsche Telekom is also working with NVIDIA, Agile Robots, Nokia and Supermicro to advance Physical AI applications on distributed AI infrastructure.
In an initial phase, Deutsche Telekom, NVIDIA and Agile Robots are testing in a lab proof of concept how AI workloads for Physical AI applications can be orchestrated for distributed inferencing between the Industrial AI Cloud and Deutsche Telekom’s distributed AI compute in the telco network edge and the on-device compute.
Deutsche Telekom is running an NVIDIA Aerial RAN Computer MGX 4U system to provide high-performance AI compute at the telco network edge, hosting parts of Agile Robot‘s physical AI use-case.
About Deutsche Telekom: Deutsche Telekom Group Profile
© 2026 Deutsche Telekom AG
