Nokia and NVIDIA have joined forces to launch the first commercial AI-native Radio Access Network (AI-RAN) platform, a move that fundamentally changes how telecom operators manage network capacity. This partnership allows carriers to significantly boost spectral efficiency without requiring additional spectrum licenses, which directly lowers the cost per bit for data transmission. Operators can now double their capacity on existing spectrum resources, making this platform critical for networks struggling with growing data demand.
Partnership combines Nokia anyRAN software with NVIDIA Aerial platform
The new platform combines Nokia's AI-native anyRAN software with NVIDIA's Aerial AI-RAN platform to create a unified solution. This integration brings NVIDIA CUDA and AI capabilities directly into the baseband system, effectively transforming the traditional RAN into a global-scale AI computer. The architecture supports three distinct deployment paths to suit different operator needs. Existing AirScale users can add GPU-driven capacity plugins, while others can deploy standalone AI-RAN nodes or utilize COTS server solutions.
Current tests have already achieved spectral efficiency improvements of more than 20%. The companies have set aggressive targets to reach 50% improvement by 2027 and exceed 100% by 2028. These gains translate to significantly improved uplink and downlink network capacity. Nokia and NVIDIA have jointly developed an AI-native Radio Access Network platform that combines Nokia's anyRAN software with NVIDIA's Aerial AI-RAN platform to optimize resource allocation.
Pilot deployments of the AI-RAN platform are scheduled to begin at the end of 2024. Full commercial availability is set for 2027. Justin Hotard, Nokia's president and CEO, described the AI-RAN as one of the most important innovations in wireless communications in recent decades. Jensen Huang, NVIDIA's CEO, emphasized that the collaboration brings AI directly into the baseband system to transform the RAN into an AI computer with global scale.
This launch marks a shift toward AI-driven infrastructure in telecommunications. The platform aims to address capacity constraints and cost pressures facing network operators globally. Pilot deployments are scheduled for the end of 2024, with full commercial availability targeted for 2027, aiming to double capacity on existing spectrum and reduce cost per bit.
Discussion
0 comments
Log in to join the thread with a thoughtful take, question, or correction.