Google plus

Fluid Edge Themes

Nokia Bring AI To The Edge with 5G

Home  /  News   /  Nokia Bring AI To The Edge with 5G
Nokia AVA Telco AI Ecosystem

Nokia Bring AI To The Edge with 5G

Nokia Bring AI To The Edge with 5G

Back in the olden days, when we used 3G and even 4G for cellular connections, AI computing at the edge was hobbled by a simple problem with a not so simple solution: data. High latency and slow file transfer speeds made offering advanced AI solutions away from the server farm difficult. The widespread buildout of 5G is helping ease these speed and latency issues, allowing businesses to usher in a new era of artificial intelligence on the edge.

One such business is Nokia. Once known for their indestructible mobile phones with seemingly endless battery life, Nokia now focuses less on handset sales and more on providing solutions to businesses.

Nokia’s AI-based AVA QoE (Quality of Experience) tool, living at the edge, has allowed partner companies to improve their customer’s experience significantly. Two such success stories revolve around media consumption; with Netflix experiencing a 59% reduction in buffering due to AVA’s AI models, and YouTube seeing 15% fewer sessions with playback issues.

Such successes are the direct result of applying artificial intelligence at the edge. The Nokia AVA QoE has shown that at can improve real world file transfer speeds that result in a better customer experience.

According to Stefan Pongratz, a Vice President at Dell’Oro Group,

“The increased complexity with the various 5G technologies in combination with the shift towards OpenRAN will potentially introduce new challenges to CSP operational teams tasked with managing end-to-end performance.

Artificial Intelligence will play an increasingly important role managing this complexity and deliver the Quality of Experience (QoE) that consumers and enterprises demand from mobile broadband applications and latency-sensitive services.

Nokia’s approach combines centralised AI to generate network-wide insights and pre-trained models with distributed AI for real-time optimisation of the RAN.”