Axiata changes course on Open RAN

When Axiata began Open RAN field trials in Malaysia, Indonesia and Sri Lanka in 2021, it hoped to find a cost-efficient way to to extend its 2G and 4G rural coverage. What it discovered were multiple deployment challenges, according to Thomas Hundt, Group Chief Strategy & Technology Officer Axiata Group, speaking at DTW Asia 2023. And as a result, it is altering its Open RAN plans.

Hundt described Axiata’s Open RAN 2G/4G field trials, which involved a number of firsts, including the testing of satellite backhauling, as a “massive, massive learning curve.” The company's conclusion was that Open RAN does not offer a commercially viable alternative to combined 2G/4G legacy network equipment.

Among the difficulties Axiata encountered was the integration of Open RAN with existing legacy networks.

“It's an automation issue, it's an integration issue, it's an interoperability issue, it's a labor capability issue,” explained Hundt. As things stand, “we do not yet see Open RAN [being] as commercially competitive as it [needs] … to be to replace legacy vendors.”

Nonetheless the company has not given up on Open RAN and is evaluating a single layer 4G fixed wireless access [FWA] use case in Sri Lanka. It is also weighing up the use of Open RAN for 5G overlay, including for private networks. “That's one of the second use cases we are working on and we have entered into a few deals already with Rakuten,” explained Hundt.

Indeed, Axiata, which has yet to deploy 5G in any of its markets apart from Malaysia believes 5G overlay offers the most credible use case for Open RAN, notably for private networks and FWA, said Hundt, in part because of how RAN software licensing models work.

“Legacy vendors … typically they charge you for every megahertz. In Open RAN you will literally have a free rights licensing,” explained Hundt. “It doesn't matter how many megahertz you're powering up and that is where eventually, from a commercial standpoint, the business case [will] favor O-RAN.” The economics of Open-RAN would be further bolstered by global, wide-scale, Open-RAN deployment, said Hundt..

Nonetheless, Hundt stressed that the disaggregated nature of Open RAN means its deployment requires skills that a typical telco does not possess in-house.

“The integration is complex. And without automation at large, you will not be able to make this work.” For this reason, Axiata DigitalLabs is developing capabilities to support Open RAN deployment, including an automation engine, so that, longer-term. Axiata can scale up Open RAN across the six markets in which it operates.

Integration, however, is not the only obstacle currently standing in the way of widespread Open RAN deployment.

“The radio performance is not comparable to fully designed legacy radios,” said Hundt.

He also noted that Open RAN needs further standardization. “If the telcos have to integrate the radios ... and there is no designed compatibility, the effort to interoperate is going through the roof and that's challenging.” Nor does he believe Open RAN can currently support a carrier-grade service. “We probably had 70% or 80% [of carrier grade service], but we are serving real customers,” he explained. “We cannot let them down because we … have a 20% gap to close.”

At the same time, Hundt recognized that Open RAN software offers CSPs agility and is evolving extremely quickly. But he believes traditional vendors remain very much in the race.

“We still need to solve the business case and we should not forget that the legacy vendors are not finished. They [continually] develop, they upgrade their equipment, so it means O-RAN has to literally develop twice as fast as the legacy to catch up.”