T-Mobile has completed what it, Nokia and Nvidia claim is the world’s first AI RAN voice call, from a commercial cell site in Bellevue, Washington, where the carrier is headquartered. The call ran over T-Mobile’s live production network using a Nokia Massive MIMO radio, Nokia 5G anyRAN software and server infrastructure built on Nvidia’s Aerial RAN Computer platform, connected to a commercial 5G core and commercial user equipment, Nokia said in a blog post.
The significance the vendors claim is the setting. Previous AI RAN work happened in controlled lab environments. This call used real spectrum and commercial network infrastructure, which moves the technology, in Nokia’s framing, from demonstration toward deployment. The next phase of field trials will extend to new AI for RAN algorithms and AI workloads running alongside the radio functions on a shared accelerated computing platform, including additional spectrum bands and inferencing workloads.
AI RAN represents an important evolution of the radio network, said Pallavi Mahajan, Nokia’s chief technology and AI officer. Nokia’s vision is to evolve the RAN into an increasingly software defined, AI native platform, creating a common foundation for future AI enabled services and the evolution toward 6G, with an ambition of two times spectral efficiency, putting more capacity from the same spectrum at the centre of the design.
Nokia’s first commercial AI RAN products are not expected until next year, and T-Mobile is one of a long list of carriers working with Nokia and Nvidia on the technology, including SoftBank, BT, Vodafone, Orange, Elisa, e and others. The Bellevue call does not settle whether operators will buy GPU based radio networks at scale. It does remove the easiest objection, that the architecture only works in a lab.