This week in Shanghai, the 3GPP concluded that the specification of Release 18 can be considered ready and stable to be “frozen.” This marks a major milestone for 5G-Advanced since now after three years of specification work, the handset manufacturers and network vendors like Nokia can proceed to start selling 5G-Advanced compliant solutions to the market. With the specification frozen, all players in the ecosystem can trust that potential future corrections in the specification will have no impact on the handsets that follow the June 2024 specifications.

We have been driving the specification work

Nokia has been one of the key 3GPP contributors in pushing the work forward on 5G-Advanced. We have provided major contributions to various instrumental features, including enhancements for enabling efficient extended reality (XR) use and Immersive Voice and Audio Services (IVAS) codec support that create a spatial and improved listening experience. We also drove the specifications to enhance RedCap, we improved the uplink coverage and specified the timing as a service capabilities. Nokia led the specifications to enable flexibility to allow operation in dedicated spectrum allocations with bandwidths between 3 and 5MHz.

And Nokia worked on network energy savings by defining a more dynamic and granular control of energy use of the base station under low-to-medium loading scenarios.

Lessons learned for future releases

As with each release, we continue to improve the operational procedures within 3GPP on how we come to a consensus on a new release specification with a diverse set of members.

I’d like to summarize my learnings in three main points.

First, the general sentiment among the delegates with whom I spoke this week was that the specification quality is now clearly better than it was at the time of freezing of previous 5G releases. This is because, compared to previous releases, we had foreseen three additional months between “specification creation” and “specification freeze.” We used this additional time to cross-check the specifications, so that as little as possible mistakes would end up in the final frozen release. In this way, we have ensured that the quality of this release is higher than ever before.

Second, having 400 optional features in the physical layer for the user equipment (UE), such as different types of channel feedback, is way too many, as it complicates the protocol side specifications and easily causes errors in specifications. At the same time, it is obvious that toward the 6G era we will likely need a different approach in 3GPP, as splitting all the possible capabilities into the smallest possible level comes with a high likelihood of market fragmentation. While chipset manufacturers and network vendors might be talking about the same high-level feature, in practice they might select another subset of the same high-level feature, resulting in an implemented feature that could not be activated. 

Third, we do not always need to reinvent the wheel. We should leverage more legacy capabilities that may be present in other technical specification groups. In this way, there were some “bonus” features in Release 18 that came at the last minute. In December 2023, I took part in an off-line session in Edinburgh focused on a new capability to reduce handover interruption time, targeted for Release 19. This so-called RACH-less handover is important in XR use cases and for industrial automation, where the application cannot tolerate tens of millisecond gaps in the data flow when connecting to another antenna. But the conclusion was that it was too challenging to achieve due to the required work, so we decided not to pursue it in the next release.

However, I learned that all the necessary elements for support of a RACH-less handover operation had already been completed in different topics, worked in parallel sessions. I found part of the missing pieces in the Non-Terrestrial Network work item while the rest of the missing pieces had been done in parallel to the work being done for the mobile Integrated Access Backhaul (IAB). So 5G-Advanced Release 18 is now even more capable than we expected at the start of the work.

Moving forward, I’m confident that we can discover more hidden gems by applying AI technologies during our specification work in 3GPP. One day, I may even be able to ask ChatGPT how to create a specific new functionality based on existing components.

5G-Advanced 101

More information on the key features of 5G-Advanced have been described by my fellow Nokians over the past three years in blogs and whitepapers, starting from the first impressions of Release 18 and actual approval of Release 18 packages back in June 2021 and December 2021. We have created a training course for those that wish to better understand the details of the new capabilities of 5G-Advanced and, together with our partner TechInsights, we did primary research on the business opportunities. Three years after the first workshop, which was held electronically due to Covid, the first 5G-Advanced Release specification is now complete. After implementation, consumers and enterprises will be able to experience and enjoy the exciting new capabilities of Release 18, while operators will be able to reduce the costs associated with operating their networks.

Next week, I will begin to concentrate solely on Release 19 and beyond, as we again pursue thrilling new ideas to explore the bridge to 6G.