O2 and Nokia’s AI Trial Lifts 5G Spectral Efficiency by 20%

O2 and Nokia have completed what they state is the UK’s first live deployment of an AI-powered RAN feature, boosting spectral efficiency by up to approximately 20%.

The deployment took place on O2’s live 5G network at its Giga Site in London Paddington, serving customers at one of the UK’s busiest transport hubs.

The project is part of O2’s £700m (US$932.4m) Mobile Transformation Plan, which aims to improve network reliability and performance with smarter technology and targeted investments.

In this trial, we saw measurable improvements in network performance and spectral efficiency without any material increase in energy consumption. We also observed no adverse impact on latency, while maintaining network accessibility and reliability throughout.

Jeanie York, Chief Technology Officer at O2

How AI optimises mobile sites

Nokia’s solution represents a shift from the traditionally static algorithms used to manage radio resources in mobile sites.

The technology takes a more dynamic approach, using predictive AI embedded within the radio network to continuously learn from changing conditions.

It analyses network traffic and anticipates shifts in demand, enabling the site to automatically adjust how spectrum and capacity are allocated without manual intervention.

Driving spectral efficiency gains

During the trial, the feature’s ability to automatically optimise radio resources improved spectral efficiency by approximately 20%.

This allowed the Paddington site to carry more data using its existing spectrum and network infrastructure.

The trial also recorded significant throughput gains during periods of high demand while maintaining stable network accessibility and reliability for customers.

Key facts

• Technology: Nokia’s AI-powered RAN software.

• Location: O2’s live 5G Giga Site in London Paddington.

• Key Result: Spectral efficiency was improved by up to approximately 20%.

• Mechanism: The feature automatically optimises 5G radio resources based on learned and anticipated network demand.

• Outcome: Increased data capacity on existing spectrum and infrastructure, with higher throughput during peak demand.

• Wider Context: Part of O2’s £700m (US$932.4m) Mobile Transformation Plan.

Better experience and future growth

The results demonstrate how AI can help O2 manage growing mobile traffic more efficiently.

This supports faster downloads, higher-quality video streaming and reduced buffering for customers in busy locations where demand can fluctuate throughout the day.

“Mobile data demand continues to grow, so we need smarter ways to make the most of the spectrum and infrastructure we already have,” says Jeanie York, Chief Technology Officer at O2. 

“This live trial with Nokia showed how AI can improve network efficiency and help us respond more effectively to changing customer demand. 

“By making better use of existing network resources, we can deliver a better experience for customers while supporting future traffic growth.”

In response to questions from Telco Magazine, Jeanie adds: “There is a strong link between spectral efficiency and energy efficiency. By using AI to optimise radio resources in real time, we can make more efficient use of available spectrum and create the potential for energy savings over time. 

“In this trial, we saw measurable improvements in network performance and spectral efficiency without any material increase in energy consumption. 

“We also observed no adverse impact on latency, while maintaining network accessibility and reliability throughout.”

Jeanie says that no additional hardware was required for the trial and that the AI capability was deployed using existing network infrastructure.

“The successful results from this trial reinforce the role AI can play in building more efficient and intelligent mobile networks,” she adds. 

“We intend to deploy the technology more broadly as part of our wider network transformation programme.”

AI moves from lab to live

The trial marks a notable step in applying AI to live commercial networks rather than keeping it in experimental stages.

“This milestone demonstrates how AI is moving from experimentation to delivering measurable benefits in live mobile networks,” says Emma Falck, President of Mobile Networks at Nokia. 

“By embedding intelligence directly into the RAN, operators can make better use of existing spectrum resources, improve network performance and create a stronger foundation for future AI-driven services. 

“Our collaboration with O2 shows the significant potential of AI-native networks to enhance customer experiences while improving operational efficiency.”

This announcement follows a recent update from Nokia and NVIDIA on their AI-RAN progress and a wave of similar news from operators around the world.

Virgin Media O2’s cable network upgrades

Following on the heels of this news, Virgin Media O2 has announced the rollout of its 2Gbps broadband services over its gigabit cable network. 

Upgrades are complete in 18 locations and the internet service provider (ISP) expects to expand the service to the majority of its fixed broadband network in 2027. 

The upgrade means the Gig2 service is now available in areas served by Virgin Media O2’s DOCSIS-based HFC network. 

Together, these areas cover around three million homes. Combined with the areas passed by Virgin Media’s fibre network, which also supports the Gig2 service, more than 10 million households can access the ISP’s fastest broadband service.

Some of O2’s key partners
 

Ericsson: A primary network partner supplying Virgin Media O2 with both 5G RAN equipment and the core network technology that powers its 5G services.

Nokia: A major 5G RAN vendor for Virgin Media O2, deploying radio equipment across large parts of the UK to deliver 5G coverage and capacity.

Samsung Electronics: A strategic partner selected to supply virtualised 4G and 5G RAN (vRAN) solutions, powering Virgin Media O2’s commercial Open RAN network rollout.

Cisco: This company provides key components of Virgin Media O2’s core IP network and automation software, managing data traffic across the mobile and broadband infrastructure.