Plans from the latest 3GPP meeting indicate a shift towards the direction favored by operators, particularly regarding the 6G air interface. This means that operators are less likely to have to replace RAN equipment. These plans are in the early stages, but they show a sense of willingness to keep things simple and a softening of the more extreme ambitions for 6G standards.
We noted in our coverage of the 3GPP’s 5G specifications, many years ago, that some mobile stakeholders were frustrated that the first wave of standards stuck with OFDM (orthogonal frequency division multiplexing) in the air interface, rather than adopting a more radical modulation technology.
Those players will likely be disappointed again by the recent news that the 3GPP RAN Working Group responsible for deciding waveforms for the 6G air interface has chosen OFDM again.
The air interface is the connection between the mobile device and the active base station, which changes as the user moves. OFDM splits a high-speed data stream into multiple slower sub-streams, each transmitted over a different subcarrier frequency. These subcarriers are orthogonal to each other, which means they don’t interfere with each other but overlap in frequency.
While the 3GPP decision includes the option to keep open the possibility of enhancements to ODFM, this puts to bed the idea of an AI-native end-to-end air interface, which was one of the hopes for 6G. There may still be AI components of the air interface, such as for Channel State Information, but an AI-native interface is off the cards.
Various modulation and coding schemes were put forward for the 6G air interface. Some proposed Orthogonal Time-Frequency Space (OTFS), which creates a delay-Doppler radar image of a transmitter’s surroundings, by which software can predict the transmitter’s movements.
When combined with machine learning, OTFS can be used to predict how a transmitted signal will behave in the face of reflections and other impediments. This is the clever work of Cohere Technologies, which has an OTFS modulation scheme that sits on top of OFDM and boosts data rates and reliability.
Other methods, such as Sparse Code Multiple Access (SCMA) and Orthogonal Sequence Division Multiplexing (OSDM), have been put forward over the years.
Some comments on the LinkedIn post announcing the decision give a picture of the disappointment felt by some: “Uy…very conservative decision,” wrote one commentator; “here we go again,” wrote another. “On one hand 0% innovation. On the other, 100% chance of easy implementation,” said another.
The operator perspective
The vanguard wireless innovators will be disappointed, but the operator community will be pleased. The 3GPP decision means that they will not have to invest in a new physical air interface and the site upgrades that this requires. This has been the clear message from the operator groups, the GSMA and the NGMN in recent months.
In June, the NGMN set out its goals for 6G, saying that 6G standards must learn from the mistakes of 5G and provide a clear commercial imperative for any new technology, and a guarantee that it will draw demand from customers.
“The industry needs to develop solutions that have tangible pull from potential customers, as there is increasing concern from operators about the affordability of investment in networks for the sake of technology development,” the letter said.
The financial concerns for operators were echoed by VMO2’s Director of Mobile Access Engineering, Robert Joyce.
“There’s always lots of hype over the next generation, which generally only benefits two people: The RAN vendors and the handset vendors.”
“We, the operators, still end up getting a similar or reduced ARPU for typically all-you-can-eat data tariffs. Therefore, what’s of interest to us is cost per bit and how we can reduce the capex and opex cost of our networks,” said Joyce.
The decision to keep the same air interface made things more straightforward for the operator, Joyce said:
“6G will most likely be based on 5G freq/time domain grid. This makes it really easy to partially re-farm bands from 5G to 6G and gradually introduce 6G on existing 5G infrastructure. With declining ARPU across the industry and no obvious killer monetization use case from 5G, operators are cautious to spend loads again on 6G.”
Operators not only want to reduce any upfront costs related to building 6G networks, but they also want to make sure that their network is as cost-efficient to run as possible. The current plans for 6G make this look likely, according to the VMO2 engineer.
“6G for me will unlock new spectrum (6 GHz) efficiently. It will allow us to go further in existing (700 MHz) and new low bands (600 MHz), it will reduce cost per bit, it will save power both in the network and handset,” said Joyce.
But the latest plans at the 3GPP do not necessarily mean that there will be no 6G capex costs for operators. There are still plenty of opportunities for 6G standards to necessitate hardware upgrades.
“A waveform change is not the only thing that could necessitate capex spending, e.g., site upgrades for the existing 3.5 GHz grid will be inevitable to add support for the 7 GHz range in the future, which will likely consist of new massive MIMO panels with an increased antenna count,” said Paul Harris, Principal Wireless Architect in the CTO Office at Viavi Solutions.
While operators are not totally in the clear on new RAN spending in 6G, there can be a bit of a sigh of relief as the “5G on steroids” vision of 6G discussed in recent years is on standby.
According to the author and former Ofcom Director, William Webb, this is a good thing, since the vision was a repeat of the 5G “build it and they will come” mistake where operators invested in technology that has not paid off.
Joyce also refers to this 5G oversell from the vendors: “I’m sure someone will tell us all the cool use cases it will deliver, but those that they proposed for 5G but haven’t come yet, so they’ve slipped them to 6G.”
One point which may raise eyebrows is that one of the key features of 6G, Integrated Sensing and Communications (ISAC), will be harder to implement because OFDM waveforms are not well-suited for sensing.
“OFDM waveforms are far from ideal for sensing, suggesting that the performance of the sensing element of ISAC may be compromised, and for many users, dedicated solutions like lidar might be better,” said Webb.
This is just an early step in 6G plans, but the news will be comforting for operators who are keen to keep capex spending down.
The latest 3GPP meetings were held in Bengaluru, India, in August. These were the first meetings on 6G standardization under Release 20.