The 6G vision needs a Reset

I used to say that #5G is "just another G". Actually, I was wrong. Objectively it has been an expensive failure, rather than a valuable upgrade. That's a very different place to where we were with 4G/LTE in 2014.

I don't want to review everything that's gone wrong. I've done that before, in one of my best-read posts and slide decks of the last year or so. Poor coverage, lousy expectation-management, silly headline use-cases and the difficulty of getting to 5G standalone for all the "good stuff" have all been contributors.

What I want to do is stop 6G making the same, or similar, mistakes. After all, as much as I like sniping at bits of the wireless industry, it pays my bills. But after following the evolution of 6G for several years, I've got a sinking feeling.

On one hand, we have the "turn the dial to 11" rhetoric from vendors and bits of academia. Cyberphysical systems, 100Gbps, terahertz frequencies and even - I kid you not - "the Internet of nano-bio-things". I've written before about the near-irrelevance of XR - it's in the same overhyped mould as autonomous vehicles and robotic surgery "useless cases" were for 5G. At one level it's understandable - there's a desire for an equipment upgrade cycle, plus lots of new IPR or PhDs. Fast development is good, for obvious reasons.

On the other hand, most of the major MNOs are calling for a slow, gentle upgrade for 6G, almost entirely based on software, rather than anything that might involve spending money on networks. Using less energy would be nice, and if the satellite folk want to spend their money on capacity and coverage, then that's OK to integrate. Ideally, it would happen later rather than sooner, say 2032-2034. And it should preserve the industry status quo as much as possible, rather than heralding anything that might alter the MNO-centric model of wireless.

Meanwhile, in the real world, the 6G train has left the station. The International Telecommunication Union has put out its framework document (more on that below), the 3GPP has started early studies, and everyone and their robot dog has a timeline chart for Powerpoint.

What's less clear is the train's destination, or how straight the track to get there is. Will 5G Advanced be a convenient stop, or is it at the end-of a branch line to nowhere?

The 6G Reset initiative

This is why three of us, William Webb , Geoff Hollingworth and myself, have launched 6greset , which has an associated discussion forum and hopefully soon a lot more. (No, we haven't got t-shirts and mugs printed yet).

We want to form a grassroots group dedicated to saving 6G from itself. Which wants to try to help the industry - and policymakers, users, academia and other stakeholders - towards a more useful 6G. One which improves connectivity for all, and which actually maps to the needs of the various groups, rather than over-engineering and under-delivering again.

We believe it is time for a reset – a moment in time that causes those involved to stop, re-think direction, and hopefully develop a better vision for 6G that aligns with what consumers, businesses and public sector users want while being profitable for all of those in the value chain.

The initiative is anti-corporate. We speak as individuals, not representing employers or clients. It's not sponsored. We'll likely irritate some, bore others, and enthuse a few.

It's also very much following the Geek Way concepts from the book I wrote about recently. We're going to be fast-moving, open and take individual ownership of things. Myself, William and Geoff agree on some things and disagree on others. That's fine. I'm not getting approval for this article, or for other things I write or say. We might pivot if needed.

If you want to see how it develops, follow the 6Greset page. If you want to participate - but observe some principles like discussions under Chatham House, apply to join the group.

Reinventing the wheel

Everyone who's spent any time on 6G over the last year recognises the IMT2030 wheel, with its embedded hexagon shape. It comes from an ITU-R document by Working Party 5D (which covers mobile "IMT" systems). Called M.2160 or "Framework and overall objectives of the future development of IMT for 2030 and beyond" it is basically the central "official" vision document for 6G.

The centrepiece is called the Wheel, which includes the original 5G triangle (eMBB / mMTC and URLLC) expanded to a new hexagon adding new concepts like sensing.

ITU is currently trying to work out how to define KPIs for all the various usage scenarios (like "ubiquitous connectivity") and also capabilities such as AI integration. That's fairly strauighforward - if arbitrary - for things like throughput speeds or latencies, where a hard number can be published. It's harder for qualitative things like security.

Speaking to people involved, there's a lot of politics (and geopolitics) behind the scene on this. ITU is an UN organisation, so national governments get involved, arguing for their interests (which includes their companies, and their societal preferences).

Apparently there was a huge amount of work undertaken to get to consensus on the wheel.

Which is why I'm going to reinvent it. I'm not a fan of consensus, as it usually means compromise and a lowest-common-denominator.

So I have unapologetically and unilaterally decided to score the various capabilities and intentions. I've rated them as Refine (good, but needs careful definition and focus), Review (sounds interesting but may be impractical or only partly useful), and Reset (irrelevant to the "maximum usefulness" 6G mission). I've already written a recent post on why XR should be ignored as a key use-case, which got surprisingly little pushback.

This chart is a starting-point. There's a lot more to do here, and ideally I'd have time to dissect M.2160 line-by-line and mark up the whole thing, but that's for another day (or month).

At a top level, this is what seems to be the "quick takes"

Refine (green)

Ubiquitous connectivity: This is paramount. A key target should be good-quality coverage everywhere including remote / rural / indoor and ideally maritime and in the air. It's better to have decent coverage reliably, than excellent connectivity in patches. It also implies a need for shared infrastructure & spectrum

AI & communications: The 6G era is also the AI era. There will be many overlaps and integration points. We should define what "AI native" means, but it's inevitable. I'm working on this, including with Charlotte Patrick in our Disruptive 6G efforts.

Sustainability: Yes, 6G should be much-lower energy consumption in manufacture & use, and have less impact in other ways

Security and resilience: I'm not a security expert, but this goes without saying. It has ramifications in areas such as quantum-safe encryption and avoiding over-relying on GPS / GNSS for timing

Connecting the unconnected: Obviously a critical goal. This needs a variety of inputs, including satellite / HAPS coverage, but also suitable price points for networks and devices. This may also mean shared infrastructure, again

Review (Yellow and yellow / red)

Integrated sensing: This is really interesting and has potential... but may turn out to be something that sits outside 6G. I'm yet to be convinced the integration will be tightly-coupled, for instance in the same radio units. We'll want to integrate external sensors such as cameras, and also on-device sensing from smartphones and IoT, so "sensor fusion" may be more important than RF-based sensing.

Massive communication: To be honest, mMTC in 5G has been a disappointment. It's mostly just been 4G era NB-IoT, without any 5G extras. Maybe the new #RedCap IoT will change things further, but I don't see a sudden shift to trillions of cellular devices, especially given the plethora of other IoT connectivity options from LoRa to Bluetooth to Wi-Fi. I can't see the need for a 6G version, really. But maybe there's more here than meets the eye.

Hyper-reliable and low-latency comms: There isn't any proper 5G URLLC yet, which suggests that 6G HRLLC isn't really going to be that important either. We all know why it's been hard to get too - coverage, lack of 5G SA cores, poor understanding of end-to-end latency and so on. That said, there will likely be pockets of this, but localised and primarily in industry / B2B. So I'd translate this more to support of 6G private networks, rather than a cross-the-board capability.

Reset (Red)

Immersive communication: The wheel is already showing its age. It was created in an era where people thought the killer use-case for mobile broadband was the Metaverse. I've written at length why that's at best an interesting niche, not a central use-case. Arguably this name just means "even faster MBB / FWA" with a marketing veneer, but those aren't necessarily that needed either.

Ubiquitous intelligence: This one is a bit vague, but seems to imply that the future of computing (and cloud) is intimately tied into 6G architecture. I'd say that's backwards - 6G will lean on the future of cloud, not vice versa. This sounds to me like MEC-on-steroids, where the mobile industry thinks it will drive the future of distributed compute. Reminder - it won't. Compute is still mostly wireline, or local-access, in datacentres, phones, PCs or home gateways. It won't be orchestrated from a mobile core, sorry.

What else?

It's worth noting as well what ITU did not put in the wheel (although there are some mentions in the full document, for another time):

• There's not much explicit about stakeholders besides telcos - for instance the idea that MNOs, enterprises, infracos and governments should be considered as 6G network owners.
• There's no mention of what I'd call "business model neutrality" - service-based provision, dedicated ownership, community use, amenity models and so on.
• Nothing about sharing - whether that's spectrum, towers, infrastructure, platforms etc
• Not much about APIs or service exposure
• Simplicity and modularity - there's nothing to say there should be decoupling of RAN and core, or flexibility around identity or other aspects
• Convergence / integration / coexistence with other network and spectrum users

Summary

This article is a "stake in the ground". I haven't got all the answers, and neither do my 6G Reset collaborators. Yes, William has a book on his view of 6G, and I will be running a workshop on policy and tech options for 6G on November 28 (more details very soon). But the point is more to get the debate started, before the train picks up too much speed, even if we don't like the destination.

Follow the page, or join the discussion, or just comment here and at upcomig events. 6G cannot be the next 5G. The industry needs to pause, think, and reset.