Ericsson says that it has successfully tested AI-native link adaptation with Bell Canada.
The vendor says that AI-native link adaptation is the use of artificial intelligence in real time to respond to changes in a signal quality, interference, and other factors. The goals are to improve performance and improve customer experience in the radio access network (RAN).
Ericsson said that the in-field tests demonstrated an increase of as much as 20% in downlink throughput and as much as a 10% increase in spectral efficiency. The press release offered no details on the nature of the tests.
“We are proud to be the first operator globally to field test this groundbreaking technology with Ericsson,” Mark McDonald, Bell Canada’s senior vice president of networks, said in the statement. “AI-native link adaptation has the potential to revolutionize how we manage network resources, delivering faster speeds and more reliable connections.”
The technology, which was developed at Ericsson’s Ottawa research and development site, integrates AI into subcomponents. This AI-native approach enables it to operate in real time and more robustly to adjust for interference, medium, and poor channel quality and other issues that degrade service. Ericsson says that AI-native link adaptation technology is consistent with its role as a founding member of the AI-RAN Alliance.
AI is a powerful tool for network operators. Last month, McKinsey & Company pointed to four ways in which broadband providers can use AI. They are connecting new data centers with fiber; enabling high-performance cloud access with intelligent network services; turning unused space and power into revenue and creating “GPUaaS” businesses.
Network capacity will need to increase in order to take advantage of these capabilities. Lumen Technologies said in March that it and Ciena collaborated on a 1.2 terabit wavelength service trial across 3,050 kilometers (more than 1,800 miles) on Lumen’s Ultra-Low-Loss (ULL) fiber network.
Lumen and Ciena used 800G interfaces to test and qualify the services to support wavelength, Ethernet and IP services over the 1.2 Tbps single carrier channel. The press release said that the increased capacity is important as AI and next-generation applications — like AI-native link adaptation — emerge.