A 3GPP RAN Working Group meeting being held this week in India has made the first agreement on the waveforms that will underpin the 6G air interface – and they are the same ones that underpin 5G.
A LinkedIn post from Nokia’s Alva Kosasih, 3GPP Senior Standardisation Specialist at Nokia, said that the 3GPP meeting had agreed that the CP-OFDM and DFT-s-OFDM waveforms will be used in 6G. CP-OFDM and DFT-s-OFDM are the same waveforms that are used in 5G, giving 6G a distinctly 5G flavour in terms of the physical interface that will be developed between devices and network.
That means that using other waveforms such as OTFS to encompass high speed use cases or to increase spectral efficiency have been relegated for now. It also means that the concept of a new air interface that is fully AI-native end-to-end air interface has been put into the “to be discussed” folder, although AI could still be implemented to enhance the air interface in use cases such as Channel State Information, for example.
For now, and probably to the approval of the vast majority of mobile network operators, and certainly those forming public policy at NGMN, the 6G air interface will look like the 5G air interface. Operators have asked for 6G standardisation to be an evolution of 5G, and not to require deep investment predicated on uncertain technological advantage and monetisation. They also want to avoid the architectural handcuffs that 5G put them in, with its NSA to SA step change.
Nokia’s Kosasih appealed to that sentiment when wrote in another LinkedIn comment, “The main intention for 6G is to make a communication system which is a mature version of the previous one. There are too many options/features in 5G standards that are not even implemented yet. With this view, it makes sense to keep what have already worked and enhance from that basis.”
Spyridon Louvros, who works for Jim Platforms, pointed out in response to Kosasih’s post that the adoption of the OFDM waveforms is not the whole of the story. “There were also other waveforms strongly supported and presented, like ZAC OTFS,” he said.
Kosasih agreed that the meeting also agreed to a sub clause to keep open the possibility of enhancements to ODFM.
“We are keeping in mind that there are new use cases that need to be supported. That’s why we don’t close the door for anything new if it is really needed to support a particular use case.”
Pavithra Nagaraj, Founder and CEO of Paaru Wireless, a technology company currently in “stealth mode” wrote that although there are upsides in continuity, cost and backwards compatibility, adopting 5G waveforms for 6G might also be too conservative and limit long term progress.
Doing so would reduce room for disruptive innovation in waveform design, with potential missed opportunities for efficiency, scalability, and new use cases. It might also bring about a “comfort zone” mentality that may slow down long-term breakthroughs.
She wrote: “Maybe the lesson here is this: 6G will innovate, but perhaps not at the physical layer- instead, the big and boldest shifts may come from network architecture, AI-native design, spectrum utilisation, and integration with non-terrestrial networks.”
One industry participant who asked not to be named agreed that the definition of the physical air interface is not the most important aspect of 6G specification.
The issue for operators is that they don’t see the potential to be investing in a brand new physical air interface, with all the costs that brings on the site upgrades, rollouts, finding and building new sites.
Therefore the pragmatic choice is much more likely to be ‘more of the same’. And in any case, 5G RAN should still deliver vast bandwidth – more than enough to serve most conceivable use cases.
“The problem” wrote our industry source, ” is that the end-to-end system is too rigid to make the use cases suitably risk-free to allow an operator to sign up to a meaningful, contractual SLA.”
A stable air interface that doesn’t suck up all the Capex would “give the industry a bit of time to work out how to flex the rest of the network to do something like Slicing, but without a predetermined network architecture as the starting point”.
Kosasih said, “A fundamental change should not be expected since it may introduce more complicated features to support, which does not align with the main goal of 6G. Having said that, we are keeping in mind that there are new use cases that need to be supported. That’s why we don’t close the door for anything new if it is really needed to support a particular use case.”
3GPP is holding RAN Working Group meetings from 25-29 August in Bengaluru, India. The meetings mark the first discussions on 6G standardisation under 3GPP Release 20, alongside the finalisation of Release 19 specifications, which will advance the evolution of 5G Advanced. Over 1,500 delegates from more than 50 countries are participating in Bengaluru meetings.
