The 6GHz band: what's next?

Note: This article has been commissioned by the Dynamic Spectrum Alliance, based on my existing analysis and positions, which I provide both publicly and privately. I believe that shared and unlicensed spectrum access is essential, for Wi-Fi, specialised cellular deployments, satellite and fixed links and other applications.

Introduction

ITU’s World Radio Congress in Dubai concluded in December 2023. It included one of the most thorny topics in wireless communications – the future of the upper part of the 6GHz band, between 6425-7125MHz and whether it should be specifically identified for IMT technologies (5G and future 6G), or left unchanged.

The outcome was complex and nuanced. At face value it seemed as though the mobile / IMT industry had championed over their fixed broadband and Wi-Fi peers, as the band indeed gained this identification. The reality is a lot more complex, and we can expect continued debate – and probably slow deployment and adoption by all – in many countries and regions, especially Europe.

In the long run, this messy outcome may even turn out to have some positives, if it leads to better and more harmonised approaches to spectrum-sharing. On the other hand, it may entrench some technology and geopolitical divides.

The back-story here has gone on for several years at a national / regional level, before WRC in Dubai. Since 5G’s inception, the mobile industry has talked about “sub-6GHz” spectrum. It has traditionally ignored the 6-7GHz range, but at WRC-19 it obtained an agenda item for future discussion at WRC-23, focused on the upper 700MHz of the band in Region 1 and a portion of 100 MHz worldwide.

Meanwhile the Wi-Fi industry had been focused on the 6GHz band earlier. Around 2017 it started asking the US FCC for more spectrum in this range, with further outreach to CEPT[1] in Europe and other regulatory agencies in 2018. The FCC was positive and issued an NPRM[2] (notice of proposed rulemaking) about the whole 6GHz band between 5925-7125MHz in October 2018, and later opened it to Wi-Fi and other unlicenced use in 2020[3].

Numerous other countries have followed suit with unlicensed use, either for the whole 1200MHz band, or for just the lower part between 5925-6425MHz. All of this occurred outside of ITU discussions, as the band is already allocated to ITU’s mobile service and spectrum rules are adopted at the national level to avoid interference to the existing incumbent users. (Note: the ITU’s term “mobile service” is a broad category, whereas IMT is a specific application within that).

Why WRC-23 mattered for 6GHz

WRC-23 eventually considered two separate proposals for identifying IMT in this band within Agenda Item 1.2:

• 6425-7025MHz for Region 1 (essentially EMEA, plus some other countries)
• 7025-7125MHz for Regions 1 (EMEA) & 3 (APAC)

The mobile industry had been loudly advocating for 6GHz spectrum for 5G / IMT, with a variety of demand forecasts for data traffic and implied economic impacts. Some of the analysis does not stand up well to close scrutiny[4], however. Dissenting voices – notably from the Wi-Fi and satellite industries – campaigned for a decision of “No Change” to the previous rules, with a variety of arguments around the important of indoor connectivity and the difficulties of moving incumbent users.

The final outcome was mostly in favour of the mobile industry’s proposals to identify the band for IMT through footnotes, but with important caveats that suggests the reality is more of a mixed picture.

The identification for all countries in region 1 & 3 is just for the top 100MHz of the band, between 7025-7125MHz. While useful for localised capacity addition, it is not a gamechanger, especially for future national 6G deployments.

The main Region 1 proposal for the full upper 6GHz band crucially came with a footnote that recognises that this band has other uses as well, specifically WAS/RLAN (wireless access services & radio local area networks – which is mainly Wi-Fi) and that IMT usage would not have priority over incumbents.

What does the outcome mean for the wireless industry?

In practice, this means that each regulator can do what it prefers with the spectrum, providing there’s co-ordination issues to prevent interference across borders.

Many countries’ authorities have expressed an interest in making the top of the 6GHz band available for “radio local area networks” or RLANs. This principally refers to Wi-Fi but also covers Bluetooth, some unlicensed 5G variants (NR-U) and maybe DECT. Some markets such as US, Canada and S Korea have already permitted low-power indoor use, and are moving on to outdoor very low-power modes, as well as moving ahead with the use of AFCs (automatic frequency controllers) for standard power use.

European regulatory body CEPT and the UK’s Ofcom are currently exploring options for “hybrid sharing” or other coordination approaches to using both licensed 5G and other licence-exempt technologies in the band. Japan’s MIC has also expressed interest in such models.

While co-existence without causing interference is challenging, we should expect various schemes and trials to explore the options – and indeed, the experience gained is likely to be fed back to the general theme of advanced sharing technologies and business models in the 2030s. The UK’s Ofcom has already stated that it sees the phrasing as being compatible with its proposed sharing concepts.

While the Chinese market is large enough for its domestic suppliers to address it as a priority, it is less clear that the market will have sufficient scale for international vendors, especially if band-specific technology modifications are needed for sharing scenarios.

Also, numerous countries (especially in the Americas) in addition to the lower 6 GHz, have already made the upper part of the band available for unlicensed use, for instance by Wi-Fi 6E and 7. Indeed, there are probably now a billion or more devices in the market capable of using those frequencies. That means that those countries will likely never use 6GHz 5G/6G, unless there are substantial changes to the technology allowing it to be a better citizen in terms of spectrum-sharing.

Importantly, the scale effects apply globally – because chipsets and devices can already cover the upper part of the band where permitted, any changes in regulatory policy can be exploited immediately. Economic and social benefits can be counted from Day 1, so any new unlicensed model for the Upper 6GHz can “hit the ground running” in terms of rapid adoption.

The outcome (including a lack of any further agenda item for WRC-27) also appears to be a signal for other countries to continue allocation of the band for unlicensed use, particularly Wi-Fi.

Comparisons

It will be very important to see comparative results from countries which have allocated the full 1200MHz for unlicensed use, compared to those that just have the lower part of the band. There will also be attention paid to the results of AFC and higher-power usage.

While industry studies are important, regulators may pay more attention to real-world customers’ comments, such as manufacturers deploying Wi-Fi 6E/7 for industrial automation in different regions, or ISPs supplying home broadband gateways under different rules.

If there are significant reported differences, either in terms of upper band access or higher-power permission, this could impact regulatory decision-making, especially if this detail is made available in the next 12 months while trials on hybrid sharing models are still ongoing.

IMT land-grab

Following the identification of the upper 6GHz band in a number of countries / regions, there is a high probability that the mobile industry will try to encourage allocation or auction processes to occur as soon as possible.

There is growing awareness that 6G mobile in the 2030s is unlikely to gain access to large tranches of new exclusive-licensed spectrum, even in the 4400-4800MHz, 7125-8400MHz, or 14.8-15.35GHz bands (or subsets of them) being studied for WRC-27. In many regions there are important incumbents, such as NATO and other military users, which will resist attempts at clearance from the bands, even if they are identified for IMT. There is a high likelihood that most or all new 6G target bands will have to be shared in some way.

This means that there will likely be a redoubled effort to claim as much other exclusive spectrum as possible in the near term before “the shutters come down”. That is likely to focus on 6GHz in the near term, along with some other bands in the 3-4GHz range.

Given there is little near-term requirement for 6GHz for normal mobile broadband use (even for densification) and unclear roadmap for support in smartphones outside China, it is probable that these arguments will centre on fixed wireless access.

It would take years for Upper 6GHz cellular to be deployed

Irrespective of product availability, there are various other constraints on the timing of theoretical Upper 6GHz 5G in markets such as Europe, Japan and beyond. In particular, many countries have sizeable populations of incumbent users in the band, many of which are in critical sectors such as satellite communications, fixed links for utilities or existing mobile backhaul, as well as scientific passive observation.

It would likely take several years of consultation, planning, agreement on sources of funding, band clearance and then auction / release of spectrum for exclusive IMT use. Legal challenges may prolong this further.

Summary – what to expect?

Taking all this together, Disruptive Analysis’ best prediction at the moment is the following:

• China (and vendors such as Huawei) will push ahead with 5G in the upper 6GHz band, with fairly near-term deployments, although the extent of these is uncertain.
• The US and other countries such as Canada, South Korea, Saudi Arabia and many in LatAm will continue with unlicensed use, mostly Wi-Fi, in the full 6GHz band, including use of higher power / outdoor versions alongside AFC spectrum-sharing systems.
• European countries and some others will spend time further investigating hybrid sharing schemes in the upper 6 GHz, that are either dynamic, or which divide between urban / rural or indoor / outdoor profiles. One option could be for fixed-wireless use only. If the sharing concepts cannot be made to work, they will have to decide one way or the other – perhaps on a country-by-country basis. Nothing is going to happen quickly, and if macro full-power 5G is allowed, a lengthy process of clearing incumbents would be needed. Realistically there’s little chance of much 6GHz 5G use before 2030, even if rules permit.
• India’s trajectory is hard to gauge at this point. Its recent telecoms laws and policies seem to indicate that it is becoming more friendly towards IT, fixed broadband and Wi-Fi. But there is also a very significant mobile industry lobby. Making the lower 6GHz unlicensed seems probable, but we can expect a lot of political wrangling over the upper band.
• Some other countries seem supportive of upper 6GHz for 5G, but it is unclear what the use-cases will be, given its propagation constraints. It may be used for fixed-wireless access first.
• The Wi-Fi ecosystem and its adoption in the full 6GHz band will continue growing rapidly, given the scale of countries that have existing regulations in place. Others will have to consider the trade-offs of opportunity cost vs. future optionality.

Whatever the outcome, we can also expect the upper 6GHz lobbying battle to continue at a country-by-country level for the foreseeable future.

#6GHz #WiFi #5G #6G #spectrum #regulation #mobile #wireless #radio #spectrumsharing #fixedwireless #telecoms

References

[1] https://www.cept.org/ecc/groups/ecc/wg-se/se-45/client/meeting-documents?flid=8135

[2] https://www.fcc.gov/document/fcc-proposes-more-spectrum-unlicensed-use-0

[3] https://www.fcc.gov/document/fcc-opens-6-ghz-band-wi-fi-and-other-unlicensed-uses-0

[4] https://www.dhirubhai.net/posts/deanbubley_regulatory-mwc-mnos-activity-6902983611298234368-YTp8/

The analysis does not distinguish indoor vs. outdoor use, and the traffic model employed uses a multiplier called an “activity factor” with no clear justification. The economic model does not discount future values, despite any putative deployment being years away.