Review: RSPG Opinion on Upper 6GHz in Europe
The EU's advisory group opinion can be interpreted in a number of different ways
It’s not over until it’s over…
On November 12th the European Radio Spectrum Policy Group (RSPG) issued its final “Opinion on a long-term vision for the upper 6GHz band”.
This follows on from numerous meetings, consultation input from a myriad of sources, technical work done by CEPT, a variety of drafts, vitriolic debates at spectrum conferences, backroom discussions and a huge volume of articles, lobbying documents and position papers (including from me).
Note: I’m using U6G as an abbreviation
The document is obviously written “carefully”. It’s been through various drafts and clearly been crafted as an attempt to translate political pressure to something technically-defensible and which fits with all the various EU regulatory and legislative paraphernalia. (I wrote about the policy angle and especially German influence recently, here).
But the recommendation also masks the details of what legitimate evidence-based analysis yields, and instead falls back on other documents and policies as “cover”. It has some gaps and non-sequiturs, especially in how it reached the final answer.
But I wonder if there’s more to it than meets the eye, or an initial read.
The outcome has been widely interpreted as a win for the cellular / IMT camp, as it recommends that the EU dedicate at least 540MHz of spectrum for prioritised full-power mobile networks, with a caveat that it must work around some satellite and scientific incumbent users.
Reading various news articles (see also here) and industry responses (and GSMA here), there is a sense that “Wi-Fi lost”. Wi-Fi and similar unlicensed / light-licensed technologies might get the bottom 160MHz of U6G, if WRC-27 allocates more spectrum in lower 7GHz, above the mobile chunk. It might also get to be used on a non-prioritised basis in U6G, subject to more work by CEPT on coexistence. There is also some scope for national flexibility.
But is everything in the RSPG report really that clear? And do the breathless online “takes”, smug congratulations and angry responses really reflect the final U6G result?
After reading it, I came to the realisation I could interpret in three different ways:
Cynical: “We’re saying what politicians and (directly or indirectly) the mobile industry want us to say, with a small nod to the others to pretend we’ve listened”
Can-kicking: “We’re actually procrastinating, because a few things might change over the next 2-3 years, and frankly there are bigger political questions here too”
Cunning: “There’s quite a bit of wiggle-room here, but we’re going to be oblique and let you work out the loopholes yourselves, so we don’t get shouted at”
I’ll get to those. But first, a bit of a diversion about how we got here and why.
Background
It’s quite reasonable to ask why all this is taking so long, given that the US has had the entire 6GHz band decided for over 5 years now, with Canada, S Korea, Saudi Arabia and assorted other countries also deciding on the fully-unlicensed option. China (sort-of) unilaterally announced a couple of years ago that 6GHz is all going to be mobile / IMT, although may now be reconsidering the lower part of the band.
The answer is part-technical and part-political. Technically, low-power unlicensed use of the band “plays nicely with others”, and doesn’t involve much change by incumbents, and can also be done without lots of international negotiation or harmonisation.
By contrast, high-power mobile in 6GHz means lots more testing, lots more clearance or creation of “masks” or sharing mechanisms to protect incumbents, lots more analysis of demand - and then some sort of auction or allocation process.
The politics is a different matter, especially now in the era of trade wars, spectrum-as-foreign policy, and a focus on sovereignty and regional champions.
The regulatory and legislative processes for spectrum, and the diversity of groups involved in European spectrum work is very complex. Some of the agencies involved are national, some are at the 27-country EU level, and some are genuinely pan-European (46 countries).
As well as the regulatory bodies and their spectrum-advisory groups, there are also political entities (national ministries, European Parliament) and other agencies such as NATO that factor into decisions.
For U6G in Europe, the key players are:
RSPG, which is linked to the European Commissions DG-CNECT (Directorate for Communications Networks, Content and Technology). It advises on radio spectrum policy issues.
CEPT, which is the European Conference of Postal and Telecommunications Administrations - basically a grouping of 46 countries’ national regulators, which also represents Europe at the ITU. Its Electronic Communications Committee (ECC) is responsible for radiocommunications and telecommunications matters. It does a number of market and technical studies that relate to regulation.
NRAs, (National Regulatory Agencies) are the various countries’ telecom or separate spectrum regulators.
RCS (Radio Spectrum Committee) develops technical implementing decisions for the EU and Commission, and also issues mandates to CEPT
One thing to note is that none of these groups really has much exposure to the requirements of enterprise users - and to be fair, there isn’t a proper “Enterprise Spectrum User’s Association” to represent them. I suggested that in 2024.
Lots of preamble, lots of acronyms
I suspect that most people reading the document do what I did, and skip straight to section 5, which actually articulates the main Recommendations.
The other sections are mostly background to the RSPG itself and various applications for 6GHz. There’s some good stuff, some of it probably intended to highlight that the authors do understand a bunch of the issues raised, such as indoor use, new radio technologies, possible Wi-Fi and 5G / 6G use-cases and so on.
There’s also a lot of excellent background, which covers the 6GHz incumbents (fixed links, science & satellite observation etc) and that they are (a) widely varying in use between countries, and (b) are actually really important and hard to move.
A lot of the document is written in “regulatory-ese”, as could be expected for something that ultimately references legal instruments such as the ITU’s Radio Regulations (RR).
There are numerous unusual acronyms, which don’t directly align with normal industry terms, as they have to avoid brands or specific protocols as the rules are (supposedly) tech-neutral.
The key terms to understand are:
MFCN, which actually stands for Mobile or Fixed Communications Networks, but is here used to mean “high power licensed mobile use”, ie mostly MNO-grade 5G / future 6G
WAS / RLAN, which means Wireless Access Services / Radio Local Area Network, which essentially means Wi-Fi, but also implicitly includes Bluetooth, local / low-power unlicensed cellular and proprietary or non-certified versions of WLAN.
Other incumbent uses, such as fixed service (FS), fixed satellite service (FSS) and mobile service (MS), radio astronomy (RAS) and Earth Exploration-Satellite Service (EESS).
It’s a bit unclear exactly where different sorts of private mobile network fit - which is a notable omission throughout. They could be MFCN (especially if high-power for utilities or public safety), or maybe WAS. It could do with more clarity.
So what does the report really recommend & why?
The whole report can basically be distilled down to section 5.3, and paragraphs 4-7, and especially paragraph 5.
The final document has come after a public consultation to an earlier draft, which suggested a band split, and asked for feedback.
But here, it adds the line “and the preferences expressed by Member states”.
Frankly, that’s the cornerstone here.
Cynical or captured? So much for the consultation
On the face of it, it seems that RSPG has been swayed more by political influences and “the preferences expressed” by governments, than by technical analysis and the weight (and credibility) of consultation responses.
I’ve read through many of the responses, spoken to various stakeholders at this week’s DSA Summit, and generated summaries and aggregate analysis by ChatGPT Pro (although it’s terrible at generating text tables).
The Opinion flies against most of the hard evidence and consultation responses provided by the Wi-Fi, enterprise and FTTH sectors, which advocated for U6G to be either be made unlicensed to align with US, S Korea and other leading markets.
They argued on the basis of indoor wireless needs and outdoor/indoor propagation realities, existing Wi-Fi congestion, extending FTTH to end-user devies, flattening mobile demand, and the existing proven demand and product ecosystem for 6GHz. A particular need is in enterprise, where a good Wi-Fi design needs 8+ non-overlapping channels, for instance to minimise interference in an office or hotel with multiple floors. The lower 500MHz of the 6GHz band can only accommodate 6x80MHz or 3x160MHz channels.
Conversely, the mobile industry leaned on hypothetical 6G demand scenarios and use-cases, and a tenuous link between new 6G spectrum, and supposed innovation and EU sovereignty. It appears to have sold this fairytale vision to some in the political arena, as I discussed recently in a LinkedIn article.
Sadly, the RSPG Opinion seems to be built on the usual layer-cake of assumptions typical in parts of the telecom and mobile. It takes the unrealistic scenarios of the ITU’s 6G roadmap, and then extrapolates them to 6GHz.
“The RSPG expect that with 6G, advanced usage scenarios will emerge, enabled by technologies such as AI […] Potential 6G application areas range from extremely high data rates for immersive VR & AR to […] Internet of Things (IoT), autonomous driving, and telemedicine. The increasing adoption of, digital twins, and real-time industrial automation, will further drive demand for […] ultra-low latency supported by e.g. edge computing. Smart city infrastructures, enhanced public safety networks, and next-generation satellite-terrestrial integration will also require additional spectrum resources. Furthermore, future networks will need to accommodate new forms of massive machine-type communications (mMTC). All these new use cases will intensify the need for spectrum.”
Yes, we’re back to 5G-era hype, warmed over again, as justification for more spectrum. As a reminder, cellular IoT accounts for <1% of mobile traffic, and autonomous vehicles don’t generally rely on 5G. Nobody wants “cloud braking”. VR/AR is mostly indoors on Wi-Fi, and the “offload to mobile edge AI inferencing” is wishful thinking at this stage, mostly from the GPU-aaS zealots.
Some of the consultation responses were from incumbents, especially those using fixed microwave links, or using the U6G for scientific or earth-observation purposes, such as radioastronomy and sea-surface temperature measurements. To a fair extent, their concerns have been recognised by the RSPG as legitimate and worth of protection.
But if view this report through the “cynical” lens, maybe it was just a face-saving exercise, and the real decision was wielded by the various governments, and especially the recent German flip, plus maybe the homes of a few of the big MNOs?
Can-kicking? Is RSPG delaying a real decision?
There are two other reads I could make.
One is that RSPG needed to put something out now, and perhaps “preserve optionality” until a lot of other things become clear.
A lot of important events are going to occur during 2026 and 2027:
The European Commission Digital Networks Act, DNA, is due for (delayed) publication in January. It is expected to amend the EECC and may include views on spectrum harmonisation, as well as thoughts about network deployment and priorities such as both 5G and FTTH uptake and use, and perhaps views on 6G.
The EU’s connectivity targets in the Digital Decade Policy Programme 2030 (DDPP) are due for review and possible revision in mid-2026. The DDPP is referenced in the RSPG report, so obviously any changes could impact the calculus underlying 6GHz strategy. Who knows - maybe it will finally add a KPI for user-experience at the device, rather than just counting fibre/gigabit to the front door of a house and the “network termination point”?
The RSPG is itself working on a 6G Spectrum position for publication in 2026. It put out a 6G "Strategic Vision” paper in February 2025, which echoed some of the ITU’s more fanciful ideas about, but added some notes of realism. Oddly, the U6G paper seems to be a bit less skeptical and is more definitive that “some usage scenarios of IMT-2030, e.g. ‘Immersive Communication’ and ‘AI and Communication’, will require mid band spectrum”. Hopefully the 2026 document will come back to earth and either espouse the NGMN’s 6G vision of “software upgrade please”, or that of the 6G Reset initiative that I curate with William Webb & Geoff Hollingworth.
The report specifically references WRC-27 at the end of 2027, and adds a rather weird dependency on the outcome of the agenda item on 7.125-7.250GHz. Basically if that is not allocated to IMT, it will flip to supporting IMT in the bottom chunk of U6G in Europe instead. It wants >600MHz one way or another, enough for 200MHz per MNO in a 3-operator market.
European regulators are keen to watch what happens elsewhere in 2026-27 with 6GHz for both cellular and Wi-Fi. The Hong Kong auction was a damp squib, and there’s not a heap of enthusiasm elsewhere. India and a few Gulf states are looking at it, but even China hasn’t actually allocated the spectrum yet.
They probably want to see more data on how Wi-Fi uses the lower 6GHz in Europe, and perhaps compare how it works with parts of the world where the full band is available. There’s also not much data on either AFC-enabled standard power, or the proposed VLP very-low-power variants yet.
Further complicating the U6G process is the fact that there is already a Dec’2024 mandate to CEPT about spectrum sharing that is still being worked through. “to study feasibility of and develop least restrictive harmonised technical conditions for the potential shared use of the 6425-7125 MHz frequency band [paraphrased: by mobile & Wi-Fi type technologies]”.
It has three tasks as part of this mandate:
Study and assessment of coexistence and compatibility of either cellular or Wi-Fi with incumbent spectrum users. (final results: March 2026)
Study of feasibility and scenarios for the potential shared use between cellular and Wi-Fi (final results: November 2026)
Development of harmonised technical conditions. (final results: July 2027)
In other words, it’s going to be late 2026 before Europe decides if cellular and Wi-Fi can coexist, and late 2027 to decide how that would work. Then there’s WRC-27 in late 2027 which will determine the lower-7GHz situation, so basically it’s going to be early 2028 before there’s any final clarity about the band.
So it’s quite possible that RSPG has decided to do something that they know won’t actually result in any auctions or deployments for quite a while. They give a nudge to the mobile ecosystem to develop some products “just in case”, and keep the Wi-Fi industry in a holding pattern.
This would also be a way to dodge some of the wider ramifications about geopolitics, sovereignty, or even the evolution of fixed broadband which is outside the RSPG’s wheelhouse, even if it ends up as FTTH+Wi-Fi. While it is working on a 6G vision, RSPG doesn’t have its own “FTTH and Fixed Broadband Strategic Vision”, or “Enterprise / Industrial Systems Strategic Vision” to allow it to balance requirements of fixed vs. mobile vs. enterprise wireless connectivity in terms of spectrum
In other words “This is political, much broader than just about spectrum and above our pay grade - someone else needs to make a decision”.
Consider the “Can kicked” scenario - either in terms of timing or scope.
But maybe it’s actually quite cunning?
Finally, there’s the possibility that everyone has missed the nuance.
I could argue that the document is actually a clever chess-match play, and that RSPG cunningly leaves quite a few areas of wiggle-room that could lead to a very different outcome in 2030 to that which the last week’s headlines suggest.
The spectrum boffins might still be two steps ahead of the clumsy politicians, and could just be letting people work out how to end-run the headline recommendations themselves.
In a sense, the timing issues above are part of this, but there’s more:
Maybe the rest of the regulatory process could run against the opinion & recommendations? There’s a few ways that could happen.
There’s still a good argument to have rethink on the actual use-cases and demand patterns, which might inform both top-level policy and country-by-country rules and timelines
Working with / around incumbents might raise more obstacles to normal IMT auction & macro deployments, which could also mean more scope for unlicensed as “non-prioritised” but still useful in many locations.
Sidenote: I’m not sure if these are deliberate hints at workarounds or possible winning arguments. Or if I’m just spotting the loopholes myself.
Leaving regulatory wiggle room
The first and most obvious thing is the line “In determining any preferred scenario(s) under task 2, the CEPT shall take due account of this RSPG Opinion on the upper 6 GHz band”
Firstly, “take due account” is legalese for saying “you have to pay it proper attention, but you don’t have to actually follow the recommendations if you’ve got legitimate criticisms or other ideas you can justify”.
Given that CEPT is not an EU body, but gets “mandates” from the Commission to formulate technical approaches, it can quite realistically say “we disagree, because…”.
Secondly there’s the mysterious plural for “preferred scenarios”, which gives a lot of scope, for instance different coexistence models in urban / rural or indoor / outdoor settings.
Then there’s a nod to very-low-power unlicensed as a possible tenant in U6G, even if the low-power indoors is not available. It recommends “a detailed analysis of the possibility of WAS/RLAN VLP”.
Lastly, there are various sections which hint at national flexibility - which could mean a general EU-level recommendation or decision, but considerable flexibility in each country to respond to mobile (and other) demand.
Room for reinterpreting the shape of demand
The other key area for shifting the narrative for U6G is the scope left for revisiting the demand patterns and assumptions. Industry participants should focus on this.
For instance, there’s a critical line in the report: “In the past, new bands had been identified for each new mobile generation of IMT. Whether this should continue in the future needs further investigation”.
This presumably means we need to wait-and-see for RSPG’s upcoming 6G Spectrum report, which might have another round of consultation associated with it - and scope for political input. It's a surprising lack of certainty which seems to contradict other sections.
I’m certainly skeptical that 6G needs any new full-power nationwide spectrum, given that we know 5G traffic growth is tailing off, we’re getting much more spectral-efficiency from networks in multiple ways, and ongoing consolidation of MNOs (& infra-sharing) should mean existing IMT allocations get divided 3 ways instead of 4, or 2 ways instead of 3.
In other words - if a further compelling argument can be made that a new band does NOT need to be identified for 6G, then a lot might change. My view is that argument is looking stronger by the day - speculative handwaving about 6G AR/VR or edge-inferencing offload is not a convincing reason. Let’s see some evidence.
Another interesting line is “all populated areas should be covered by a next-generation wireless high-speed network with performance at least equivalent to that of 5G”. Now, the last time I checked, “populated areas” were generally on the inside of buildings, not the outside. A clever campaign could try to get more indoor-centric obligations attached to the final rules of the band - for instance, in any future harmonised auction conditions.
Finally, this line leaves a few things unspoken: “the merit of enabling WAS/RLAN deployment in specific cases where WAS/RLAN capacity needs would benefit from using the full upper 6 GHz and where no MFCN is expected to be deployed, e.g. a large factory in rural areas”.
In other words - if nobody expects 5G/6G deployment in a given location, the Commission, NRAs and CEPT should bear in mind the needs of others. Maybe there are use-it-or-share-it options to consider here.
Don’t forget the incumbents
A final angle entirely is around the protection of incumbents in the 6GHz band. There’s a lot of emphasis in the RSPG opinion on fixed satellite, fixed microwave links, scientific observation and so on.
It notes that 20 European countries have reported 12,500 fixed links in U6G, in some cases forming parts of their public safety infrastructure. Those will be hard (and slow) to move.
The report states “the limits of national flexibility regarding the current incumbents’ use of the band” and also “how to address the protection and future development needs of incumbent services and applications”. That phrase “future development” is important - it recognises that several of the existing users groups may actually grow, rather than shrink away or be easily cleared.
It notes “the usage of the upper 6 GHz band for radio links in Sweden and Italy is extensive and is unlikely to decline in the near future… In fact, the number of links may increase”. It also highlights an “increasing trend in use” of 6GHz links in Bulgaria, Croatia, Moldova, Netherlands, and Türkiye.
In other words - it’s not just the Wi-Fi industry that will push back on U6G spectrum being auctioned. There will need to be protections for various incumbents - especially as the ITU’s Radio Regulations urge states to “take all practical steps to protect the RAS from harmful interference in the frequency band 6650-6675.2 MHz” - including separation distances of up to 500km between transmitters and observatories.
Conclusions
Many people have been writing about the RSPG Opinion on Upper 6GHz as if it’s all now neatly wrapped up, with a 5-1 victory for the traditional cellular industry, despite the consultation and persuasive arguments that have been made.
I don’t think that’s necessarily true.
There seem to be three scenarios:
Cynical: U6G is a gift, because policymakers want the mobile industry to get a few crumbs, when the DNA doesn’t given them other things they asked for
Can-kicking: Everything will be clearer in late 2027 or early 2028, so this is a way to delay the final decision - or leave it up to someone with a “bigger picture” view
Cunning: Read between the lines. There’s still a few ways to work around the pressure from certain politicians.
In other words, the game is still being played. It’s just entering extra time.