stein@opensky
Just before the Christmas period, I saw various LinkedIn articles and comments around 6G and what it is really for. It is some time now since I last posted something similar on opensky.no via LinkedIn, but of course I got triggered by these articles.
I have “grown up” in the mobile operator industry – so I may have a certain bias – but as of today, I am an independent consultant working in the wider ICT industry. I will therefore try to offer some thoughts about technology, regulations and politics– and about industry competition. Refer also my accompanying article with reference to politicians wanting to “Take the digital lead”.
I have been part of the development of the mobile industry since before 2G came to market – so I have seen how 2G went from being a European standard to various Gs becoming global standards and the main means for communication and entertainment across the globe. Along this path I have also seen companies, industries and nations fighting for dominance hidden as “technology wars” – and also how the political and regulatory regimes have changed over time – and they differ !
The same old story?
When I go back and read my article from 2021 as a “grumpy old guy” on “5G or 6G? The story all over again?”, I see that most of what I want to comment on today is already commented on there. Basically, it is about whether the telco way of work (defining a new G every 10 years) is sustainable or not, for various reasons, including:
· Operating in an increasingly complex ecosystem (and who is the G really for?)
· Growing technical complexity
· Finding the use cases and business cases (for whom?)
· Conflicting business interests
· Alternative technologies
I will not repeat the details of this here, but I recommend reading the above article (as well as other articles from 2020 referred to in there, like “5G will save the world – or will it?”, “5G is all around us – but who will eventually benefit?” and “What will really happen with 5G?”). These are all 4-5 years old now, but surprisingly mostly still true.
As I also commented on in the above articles, every “odd” G is really not needed, this has more or less been true so far. It may, however, be the end of the line now. 1G was useful – but only analogue, local and fragmented, but 2G was the first digital standard with global reach. 3G was intended to be all that 2G was, just better - but did not end up being a proper mobile internet (broadband) – which 4G did. Again, 5G was intended to be all that 4G was, just better - but with a main focus on industrial applications – which, however, have not yet materialized. 6G seems to be a repetition of 5G, but with a much more unclear purpose – and with totally unclear business models.
Looking forward at 6G, my old friend Bill Best of Azenby has recently written a very good article “5G is enough thanks” on technical, commercial, political and regulatory aspects which may hamper 6G – which I also recommend reading. I would also point to a LinkedIn post by the self-proclaimed controversial and outspoken Dean Bubley on “6G’s unclear purpose”. This time, maybe the “even” G is not needed?
So is 6G needed – and is it coming?
To cut it short, 6G is coming – simply because so many people are working on it across the globe – with their own obvious business interests. Some wider questions are however: When is it really coming? Who will benefit from it? How complex will it be? … or how much money will be lost in the process?
Is it needed? It remains to be seen – and it depends on who you ask. Many HW and SW vendors have lived off the telecom operators for decades and will clearly want to have another G to sell. Academia and research communities also clearly want work to do – and regulators have got used to a licensing regime where they collect huge amounts of money from operators acquiring licensed spectrum for this purpose. The wider question is, however, whether operators are willing to pay huge sums for 6G with a totally unclear business case. The auctions for 3G some years back ended up with huge sums – and 3G is now mostly being sunset in Europe. On this note, the battle for new licensed or unlicensed spectrum in the next round in 2027 has already started. Is it the end of the line for licensed, exclusive and very expensive 6G spectrum? I expect that there will be new spectrum models, be they licensed, unlicensed, shared etc – and I really hope that all 6G players (operators - but also others) will be able to develop sustainable business models for it.
Who will then want 6G services?
For 2G what was really in demand was mobility and the ability to call and be called anywhere anytime, i.e. to speak on the phone. For 3G it was the same, but for 4G the demand was for broadband internet access. Mobility is of course nice to have but most of the internet access happens in stationary mode. The same applies to 5G and in this case the broadband is even broader – up to gigabit speeds for mid-band spectrum and many gigabits for mmWave – the only difference being that wide-area mobility is quite unrealistic for terrestrial mmWave.
Most of the above demand assumes a consumer market – while 5G as well as 6G are really designed for industry applications – which are not yet quite there even for 5G. Technically, the “real” 5G (stand-alone) has been designed for very high bandwidth, ultra-low latency and massive IoT, however, whichever way we look at it, the business models towards the industry verticals are so far unclear and unproven – and, more importantly, the various solutions may struggle with global scale – as has been a main success factor in the consumer market for previous Gs.
What are the key 6G services?
At this stage, there are very few concrete ideas on actual 6G services and who they are for. In NGMN’s first publication on 6G some years agon (2021), 6G was focused around three very high-level requirements: 1) supporting the UN SDGs, 2) offering new services – and 3) operational efficiency – and an important technical element was automation. This is of course nice but not very concrete. It is worth noticing the third element, however, of operational efficiency. As NGMN mostly represents the operator CTO community and their typical suppliers, this is natural and good – but it does not say much about the second point on key services.
The organization developing standards for Gs (3GPP) has made their initial work plan for 6G, which includes a range of study items in areas like “guaranteed services”, “resilience and robustness”, “deployment and coverage” and “security” – all about technical aspects and again, no “killer app” in sight that someone may want to buy.
Moving to the more concrete ideas on features that may result in services, NGMN points towards sensing, AI, extended AR/VR, enhanced positioning, use of network APIs etc – and also to integration with satellite networks. Use of network APIs is already today GSMA’s and the NGMN community’s approach towards services for 5G. In my view, this is an important area of focus, as long as the operators manage to engage the developer community to drive scale towards a long-tail set of network services (remains to be seen). For satellite integration, this will obviously drive towards mobility and global coverage for 6G – which may drive a certain increased global penetration and (hopefully) some revenues for operators.
Although I have lived through some phases of “competition” between terrestrial and satellite mobile communications (refer my quite old article on “Will satellite comms happen this time?”), it seems now that the terrestrial and satellite communities are in a more cooperative mode, so let us hope that this will work for 6G (or before) – as long as business models work out.
It has been stated that 6G will be the first AI-native G, which it may – but this will probably have the most effect in the area of operational efficiency.
To summarize, although there may be a lot of technical enhancements in 6G compared to 5G, the “killer apps” for 6G are nowhere in sight (yet), however, there may be service opportunities in the areas of sensing and immersive communication, and for the long-tail set of possible services based on exposed network APIs. It is, however, not clear if anything will reach mass-market scale in a consumer market (as for earlier Gs) – but we can hope.
What about WiFi?
In my “grumpy old guy” article above and also the article on “WiFi6 and 5G – Is 6 better than 5?” the question on WiFi’s role in the ecosystem was brought up – as also in Bill Best’s article referred above. In the meantime, also WiFi7 is around now. WiFi will definitively have a role in the 6G (and 5G) era – but the question is how. This is not purely a technical question but much more a political question about industry dominance and suitable business models. With this also comes the question of licensed and unlicensed spectrum (refer above). Without elaborating more here, see my summary below.
Indoor or outdoor services?
Most mobile operators have spectrum licenses with national coverage obligations. To clarify this, however, such license obligations are there normally to cover a certain number of households in the country (seen from outdoors). It is therefore focused on outdoor mobile coverage with no requirements at all for indoors. The challenges with this are: 1) that for industrial applications and 2) for very high bandwidths (which 5G has been designed for), most of these use cases are indoors (except massive IoT). See also my article from 2022 “Gigabit indoor 5G – who can do it?”.
Summary
I may jump the gun a little here without further detailed argumentation, however, my summary and recommendations on 5G and 6G are:
At this point in time, leave 6G to the standards guys and make sure 5G is a success! We don’t need another 3G with huge investments, little pay-off and a short lifetime.
6G will come – but 2030 is too early for market launches. 5G was already too complex and launched too early - and 5G SA which has all the promises in it is not even there yet 5 years after. The proper telco cycle should probably be around 15 years these days.
The standardization work on 5G is still happening and for 6G it will go on as well – but the main challenges for both are commercial not technical.
Much work is still needed on business models for industrial applications in 5G – a main challenge being economy of scale. Large corporate verticals may have specific interests and solutions in mind and they may either be willing to pay the cost or go for private 5G networks (see also my previous article on “Looking for a private 5G network?”). For the SME market, a certain standardization of solutions will be needed to reach scale – and SMEs may not even be aware of the opportunities with 5G.
Most market communication about 5G to date (at least in my country) has been about industrial applications like autonomous driving”, “remote surgery”, “drones”, “factory automation” etc. The two issues around this, however, are 1) that 5G has not been ready for what has been communicated yet – and 2) consumers don’t care. The consumers so far see 5G as “just a faster 4G”. The metaverse and immersive communication etc are not yet on their mind. Maybe enhanced consumer applications will come after some years (?), but I for one don’t want to wear goggles!
All parties (operators, vendors, regulators, industry associations etc) need to step up on relevant communication and awareness around the benefits of 5G for industrial applications – and solutions need to be relevant and cost-effective for industries (and available at the right time) – including also targeted approaches towards SMEs so that economies of scale can be reached.
To make 5G meet its promises, it has to work indoors – not only for industrial applications but also for regular consumer use. Indoor coverage with 5G mid-band spectrum and modern buildings with very good insulation require dedicated coverage built from inside. Operators do not have the funding to do this – so they need to partner with neutral hosts and customer funding to make this happen. Regulators need to know and to promote this.
For indoor use, also WiFi is a good (and already existing) alternative, however, without going into details on this there are clear benefits for 5G above WiFi – like in-built security, seamless operation with outdoor coverage, stability and more.
It is still unclear who 6G is for and what its purpose is – and who will make money from it. It may well be that operators could be unwilling to pay lots of money for licensed 6G spectrum – unless their business case is clear. 6G is still in an early phase of development, so it might become clearer with time, however, it seems to me that there are limited opportunities for economy of scale.
Regulators need to understand the various business model aspects around 6G and to consider alternative licensing schemes, with licensed, unlicensed or shared spectrum which support a sustainable business for all.
To round off, I would like to come back to Bill Best’s article and suggest that the 6G ecosystem (whenever it is launched) will include wide-area mobility through 5G and 6G, with integrated satellite communications and also WiFi for stationary use. It will therefore be necessary for the cellular and WiFi camps to find ways of cooperation (like for the satellite community). It will also be necessary for regulators and politicians to understand that not all of the business case for 6G can be assumed for the mobile operators – which must be reflected in the 6G licensing schemes.
I would also like to support the NGMN in the view that we do not need another fundamental network upgrade for 6G as we have had (and are still having) for 5G – with the comment that 2030 is probably too early for market launches. “NGMN Alliance (NGMN) believes that 6G is the graceful evolution of communication networks into the 2030s, delivering compelling new services and capabilities for customers whilst maintaining essential offerings such as voice. 6G will build on, and extend beyond, our existing 5G ecosystem to foster new innovations which deliver value to customers and simplify network operation.”
As I have been saying earlier, maybe 5G is the last G (at least as a fundamental network upgrade)?
Comentarios