This week in Incheon, Korea, 1500 people are attending 3GPP’s first official Workshop for 6G – many of them having flown straight out from MWC Barcelona. The workshop is more of a showpiece than a decision-making forum, but it is symbolic of 3GPP’s efforts to start formalising what to prioritise within R20, the first specifications release that will contain candidate technology for the ITU’s IMT-2030. In other words, 6G.
There have been 250 submissions, and speakers representing their companies will be allocated seven or eight minutes each to make their proposals for inclusion.
And yet, if you walked the floors of MWC, you’d be forgiven if you missed any mention of 6G at all.
You had to go inside Nokia and Ericsson’s booths to see 6G on their stands. Nokia was showing results of a real world trial it carried out three weeks ago in 7 GHz spectrum, using its 128 T/R Habrok antenna in the frequency. It had found that 7 GHz performance was pretty good at cell edge, delivering 1 Gbps at 1km distance. That would be good enough to be able to use existing site topographies. And also that it is OK at going through buildings, observing 375 Mbps performance at a distance of 100m, after penetrating through two glass walls and one solid one.
If Nokia was showing real world outdoor testing results, Ericsson had a fancier set-up on-booth, having brought its test 7 GHz antenna and test UE, and was also showing similar measured performance between the two – 3 Gbps, if you’re asking. There was also a mention for Integrated Sensing and Communciations.
Qualcomm had mentions of 6G on a large screen titled Wireless Innovation which contained a number of slide-decks on topics such as 6G Giga-MIMO, a lower band system design for 6G, AI-native radio protocls, and AI-enhanced wireless efficiency.
Elsewhere much of the 6G action was centred on the test company booths, such as at Keysight and Rohde & Schwarz, with demos of Digital Twins for the AI RAN and integrated communications and sensing.
Research and IP developer InterDigital was showing two demos of Integrated Sensing and Communications use cases that, in once case combined on-board sensing data with RF sensing data, and in another used RF sensing in combination with AI to recognise the presence of an intruder in a room it had built as a digital twin.
Why so quiet?
But despite these demos, there was a notable lack of hype and presence aroud 6G – probably less even than two years ago. The vast majority of stand designs, the banners and the vast majority of the fancy demos had no up-front mention of 6G at all.
So what – given the industry formally kicks off its 6G efforts next week, is going on?
Alain Mourad, Senior Director, Head of Wireless Lab, Interdigital, agreed that although he saw some 6G branding around the Halls, it was limited. In his view that’s been driven by mobile operators’ desire to move on from being held captive by the G branding, and being cautious about vendor intentions.
Mourad didn’t say as much, but if we take NGMN Alliance as representative of the operator viewpoint, operators are at least ambivalent about the idea of going all-in on a new G. They’re asking the industry to make 5G to 6G a “graceful evolution”, and not to mess up on architecture definitions (NSA, SA and the mutiple possibiities for transitions between the two) that arose in 5G as a result of a rush to commercialise.
6G technologies narrowing
There has also been a tuning process on the proposed technologies for 6G. The fanfared use of THz and sub THz frequencies for massive near-distance bandwidth and throughputs has subsided, probably informed by the lack of success in mmWave bands. Instead focus is shifting to FR3 frequencies in the lower bands such as 7 GHz – witness the Ericsson and Nokia demos. And the RIS – Reconifigurable Intelligent Surfaces – is still present as a concept, but doesn’t need to be in 6G, according to Mourad. The perceived potential for ISAC persists, however, at least judging by the available demos at MWC this week. (See also Cohere Tehnologies on OTFS and the Pulsone)
One other thing that persists is the sense that there’s a clear dichotomy between the idea of formalising a lot of features and technology as a G, and building the sort of continuously updateable, software-based networks that operators say they are after. In other words, if you have the right platform, skills and ability to continuously iterate, upgrade and update, and a software RAN, what is the benefit of coalescing a lot of that around a hardware-led G?
On the other hand, there are still new use cases and technologies arising, whether you want to label them as 6G or not. Interdigital’s Mourad pointed out that Mediatek was showing an AI-enhanced antenna sensing capability that was in essence a 6G use case – it was just not labelled as such.
In this demo, the antennas in the phone are able to intelligently respond to whether the phone has been picked up with a left or right hand, and then adjust performance according.
ETSI CTO Issam Toufik, though, that even though 6G may not be very visible, a number of “AI” demos could also be viewed in a 6G context. And he pointed out that the idea of a software-based, AI-powered network is not disconnected from 6G, but rather integral to it. Next week’s 3GPP workshop will see AI, automation and cloud technologies proposed as a part of what 6G should look like.
Maybe the silent generation approach is a good idea. We bemoan hype in this industry. Perhaps we shouldn’t bemoan its absence.
