- Vodafone Group joined the growing list of mobile operators ready to explore AI-RAN architectures earlier this year
- Now it has opened an AI-RAN lab in Spain and its UK operation, VodafoneThree, is preparing lab and field trials with Nokia, Nvidia and Red Hat
- The UK operator sees ‘significant potential for AI-RAN’
UK operator VodafoneThree is to undertake lab and field trials of AI-RAN technology developed by partners Nokia and Nvidia in the coming months as it explores the best way to use network resources (including spectrum) more efficiently, support AI-related traffic and improve customer experience.
The move comes as part of a broader exploration of AI-RAN by the UK operator’s parent company, Vodafone Group, which had signalled its intention to work with Nokia and Nvidia earlier this year at MWC26, when the giant telco’s director of network strategy and architecture, Marco Zangani, noted in this announcement that it was working with Nokia and an “ecosystem of global partners” to advance the “development of compute-platforms at its Málaga Innovation Centre that will form the nucleus of future autonomous, AI-native and 6G-ready networks.”
That collaboration has now resulted in the opening of an AI-RAN lab in Málaga that will explore multiple technology options (for example, the use of CPUs as well as GPUs) from multiple vendor partners.
Since MWC26, many other telcos have announced AI-RAN tests and trials with Nokia and Nvidia – see Nokia bangs louder on the AI-RAN drum and Nvidia boasts greater AI-RAN and AI factory traction with telcos.
Now VodafoneThree says it will put Nokia’s AI-RAN platform – which comprises the vendor’s anyRAN software running atop the GPU-based Aerial RAN accelerated compute system developed by Nvidia (an investor in Nokia as well as the vendor’s AI-RAN platform co-developer) – through its paces, “initially in a controlled lab environment before progressing to a selected-site field trial”.
The operator will “evaluate how AI can improve network capacity, reliability and spectrum efficiency, helping deliver a more consistent customer experience, with initial use cases exploring how AI can assess connection quality and adapt data transmission in real time as network conditions change.”
Specifically, VodafoneThree will use the platform for use cases such as machine learning-based channel estimation and link adaptation. “In simple terms, these technologies could help the network more accurately assess changing radio conditions and adapt how data is transmitted in real time, helping to improve network performance and the user experience,” noted the operator, which identifies use cases such as AI-for-RAN, one of the three pillars identified by the AI-RAN Alliance.
The trials “will also examine the wider potential of AI-RAN as an intelligent, programmable network platform. VodafoneThree has identified a range of potential AI-RAN use cases spanning radio optimisation, mobility, energy efficiency, anomaly detection, coverage and capacity optimisation, quality-of-service (QoS) assurance, and the sharing of accelerated compute between RAN and AI workloads.”
The operator stated it will use the trials to explore that final potential use case – resource sharing, also known as AI-and-RAN. It noted that it will combine Nokia’s AnyRAN software, Nvidia’s Aerial RAN and Red Hat OpenShift to create “a common platform capable of supporting both RAN and AI workloads.”
VodafoneThree noted that the “trials will also explore differentiated connectivity services and shared accelerated computing infrastructure for RAN and AI workloads, bringing AI-powered services closer to customers… from streaming and gaming to working and using AI applications on the move, customer demand for differentiated connectivity services continues to grow.”
It added: “As AI becomes increasingly embedded in daily life, mobile networks will need to support growing volumes of AI-driven traffic alongside traditional voice, data and video. The programme will explore how Nvidia accelerated computing can support AI and RAN workloads on common infrastructure, helping VodafoneThree to understand how network resources could be used both to enhance connectivity and support new AI-powered services closer to customers.”
Clearly, though, VodafoneThree, like many of its peers, is yet to determine the optimum architecture for AI-RAN, in terms of just how much accelerated compute power needs to be distributed at the edge of the network to meet its own and its customers’ needs: Deploying GPUs at base stations could be challenging from a return-on-investment perspective.
But this is why operators conduct such trials – to identify the real-world advantages of emerging technologies and benchmark them against others. As VodafoneThree noted, it will compare the test results with existing network performance “to explore where AI can deliver the greatest benefits, measuring improvements across capacity, reliability and spectral efficiency, and exploring how these capabilities can be scaled across the wider network.”
But clearly, some aspects of AI-RAN will play a role in VodafoneThree’s future. Iain Milligan, the operator’s network development and infrastructure director, stated: “AI-RAN represents an exciting next step in the evolution of our network, using AI to make connectivity more intelligent, responsive and efficient. Working with Nokia, we’ll test these capabilities to understand how AI can improve network performance and, ultimately, the experience we deliver for our customers. As our 5G standalone network evolves, we see significant potential for AI-RAN to help us respond more dynamically to different customer needs and unlock new, differentiated connectivity experiences, while laying the foundations for the next generation of intelligent, AI-native networks.”
- Ray Le Maistre, Editorial Director, TelecomTV
Email Newsletters
Sign up to receive TelecomTV's top news and videos, plus exclusive subscriber-only content direct to your inbox.