New telco demand for no 6G hardware refresh will rattle vendors

Howard Watson, the most senior technology executive at the UK's BT, once developed the thousand-yard stare of men who have fought bloody wars when asked if 6G would trigger another big equipment upgrade. "We are not doing that again," he said in apparent horror at this year's Mobile World Congress (MWC). It now seems that other telcos are not up for doing it again either.

A position statement on 6G released this week by the NGMN, a club of telcos, could not make it clearer. "6G must not inherently trigger a hardware refresh of 5G RAN infrastructure," say the report's authors, who include Luke Ibbetson, the head of research and development for Vodafone, and Javan Erfanian, a distinguished member of the technical staff at Bell Canada. Where possible, they go on to say, operators must be able to introduce 6G through "software-based feature upgrades of existing network elements."

By now, traditional equipment vendors must be feeling as sweatily nervous as men about to fight bloody wars. Take Sweden's Ericsson, the world's biggest vendor of 5G technology outside China. Despite years-old efforts to kindle standalone software businesses, it still generates about two-thirds of its revenues at its mobile networks unit. Product highlights these days include slimline radio units for massive MIMO, an antenna-rich 5G technology. Promotional images typically feature women of average size carrying them like plastic supermarket bags to emphasize just how lightweight they are.

Related:6G is shaping up to disappoint, and the industry can blame itself

Ever since 2G came along, Ericsson and others have been able to rely on an equipment refresh – roughly once every ten years – for a sales bonanza. Revenues earned when telcos are extending coverage help justify the huge research-and-development costs of producing new-generation technology. At Ericsson, these came to 47.3 billion Swedish kronor (US$4.3 billion) last year, about 17% of revenues. Nokia, its Finnish rival, spent €4.55 billion ($4.8 billion), roughly 18% of sales.

Generation ex

Yet 5G has not been going as hoped, which partly explains the evolving telco attitude toward 6G. Four years since it was first launched, it has produced no uplift in sales in either consumer smartphone or business markets. An array of promised new services – surgeons stitching up a wound remotely over a 5G link, cars 5G-alerting other vehicles to the likelihood of a crash – has not materialized. That's despite the multi-billion-dollar investments in 5G that telcos have made worldwide.

Has it all been a wasted effort? Probably not. Equipment eventually wears out and needs to be replaced like old running shoes. When that happens, it's better to invest in the latest speed-boosting gear than buy more of the old. Huge volumes of mobile data traffic are now carried on 5G networks as well as more dated systems. Telcos routinely grumble about return on investment, but their critics note they are still healthily profitable on various measures.

Related:After 5G hangover, there's not much telco love for 6G

Nevertheless, some gloom has descended on the industry about 5G's perceived failures, and telcos do not want a repeat experience with 6G. It should not be a totally new tech, necessitating colossal expenditure on auxiliary systems and an overhaul of old platforms, but a fully interoperable enhancement of today's networks. Kits should only need replacing for "end-of-life" or related reasons and be an "operator-driven choice independent of supporting 6G." These, rather than anything about new 6G features, are the big messages in the NGMN report.

A pure software model

How disruptive could this be for the equipment sector? The likes of Ericsson and Nokia are already having to adapt to operator demands for cloud-native and software-based radio access network (RAN) technologies. In a traditional network, Ericsson supplies both radios and computing equipment that includes its own chips and software. In a virtual or cloud RAN, server and semiconductor specialists would provide that hardware. "It's a pure software model from our perspective," said Fredrik Jejdling, the head of Ericsson's mobile networks unit, during a conversation with Light Reading in February.

Related:6G will be held together with centimetric band

Networks obviously still need hardware. Increasingly, though, it may come from the likes of Dell, HPE and Intel, rather than the telecom-equipment vendors that normally serve telcos. And some operators might not buy much equipment at all. Under an arrangement showcased just last week, Google could bring its own servers and racks into a telco facility, using Ericsson's software to power the RAN.

It would mark a radical break with convention and herald less capital investment by operators, lower hardware revenues for Ericsson and a much bigger role for Google and other hyperscalers doing the same thing. "They are becoming a more critical path and less so the Ericssons and the Nokias," said Roz Roseboro, a principal analyst with Omdia (a Light Reading sister company).

The Nordic vendors would still make and supply the radios, of course. The challenge there is partly from new rivals buoyed by interest in open RAN, a telco movement aimed at cultivating specialists (few of which currently look very competitive). A bigger problem is the NGMN's latest demand for 6G not to require a huge swap-out of hardware in one go. It echoes comments about 6G that BT's Watson made at the start of the year. "We are going to do the same radio interface," he said. "There will be some millimeter wave hotspot supplementation rather than densification, which implies a lot of it."

Ericsson, Nokia and others could still do well as mainly software companies. While revenues may shrink, their profit margins are often fatter when telcos are making network improvements through software updates rather than building out the infrastructure. But a mainly software model would be very different from today's modus operandi. It seems they might not have much choice.