Your guide to 5G-Advanced
Podcast episode 67
The network is constantly evolving and 5G-Advanced is the next step on the long journey to 6G. Nokia’s Matthew Baker explains the full range of value this will bring – from improved consumer XR to reducing carbon emissions.
Below is a transcript of this podcast. Some parts have been edited for clarity
Michael Hainsworth: 5G is here today – sort of. While the last few years have been spent building the foundation of the next generation wireless network, 5G-Advanced will take the network beyond faster speeds and accelerate the Fourth Industrial Revolution. Nokia’s Matthew Baker points out it will deliver a plethora of innovations for enterprise, the end-user, and for network operators.
Matthew Baker: The first 5G networks were deployed already nearly three years ago in 2019, according to what's known as release 15 of the specifications. And these first 5G networks lifted the data rates and capacity for mobile broadband to new heights, together with outstanding energy efficiency as well, resulting from the very low overheads of the 5G air interface.
One of the most exciting aspects of the 5G foundation is its flexibility. It's been designed from the beginning to be highly adaptable to different traffic patterns, quality of service requirements and deployment scenarios, for example.
So, we've now seen with releases 16 and 17 that 5G foundation has been expanded to support ultra-reliable and low latency communications, sometimes referred to by the rather horrible acronym of URLLC, which provides the foundation for 5G to support industrial IoT. That's industrial automation for manufacturing, asset tracking and such like.
So, the 5G networks that are deployed today provide us with an extremely flexible foundation on which to add further new features, and as we now enter the era of 5G-Advanced, it's time to build on that with further new features that really get the most out of 5G to let it reach its fullest capabilities.
MH: So, if 5G-Advanced is made possible by the 3GPP release 18, what does that enable for the consumer?
MB: Well, it enables many things, but potentially one of the most exciting new application areas that will be particularly addressed by 5G-Advanced is extended reality or XR. Now, when we think of XR or VR, virtual reality, most of us probably think of heavy, bulky contraptions strapped onto your forehead, and we certainly don't yet think of it as something we would use every day to enhance our lives.
A 5G-Advanced network will enable XR devices to be very much smaller and lighter. So, let me explain how that will happen. 5G-Advanced networks will support new degrees of what we would call application awareness. What that means is they will be able to distinguish between different types of data that need to be transmitted in order for XR to work.
So, that might, for example, be data corresponding to your direct field of vision, directly in front of you, versus your peripheral vision, which doesn't require the same level of quality. Or audio signals versus haptic feedback for touch that has a much tighter latency constraint and so on.
Now, being able to distinguish those different data types and their corresponding quality of service requirements means that much of the processing burden for rendering XR applications can be transferred away from the user's device into the powerful computing resources of the network. And that means the devices can become very much smaller and lighter, consuming less power, and hence being able to operate with smaller batteries. So that will be a very noticeable benefit for the consumer.
But the other thing is that application awareness in the network will also increase the capacity of the network to support more XR users in a given area, and that's also, of course, very important if we're going to see XR delivering its benefits in a widespread way across society.
MH: This is absolutely fascinating to me because we were all promised one day, we'd be able to put on a pair of glasses and they would do the job of that glowing rectangle we all have in our pockets. We won't have to have a smartphone in our pockets anymore if our eyeglasses are beaming information in front of us and keeping us on top of things. And it's in that lightweight frame of a pair of eyeglasses that you just can't get right now in an extended or alternate VR reality type environment right now.
That's fascinating that the low power components vary front and center as far as the consumer is concerned. And then I think you sort of touched on something that is really important for 5G-Advanced behind the scenes for the CSP, and that's the capacity expansion that comes with that as well.
But what are some of the other behind the scenes features that 5G-Advanced provides the CSP when delivering service to the customer that the customer probably isn't even aware of?
MB: Absolutely right, increased capacity is one of the key benefits to the CSP there. But on top of that, I would add three other key areas that 5G-Advanced will bring.
Firstly, enhanced coverage, which is always important for network operators because they have to bring their services to you wherever you are. And particularly in the uplink, we see at the moment that there are challenges as new usage of mobile networks changes the balance between downlink and uplink traffic with, for example, more video traffic being generated in the user's device and therefore having to be transmitted in the uplink.
So, for example, we're going to be looking at new frequency domain shaping of the uplink signals that will enable higher uplink data rates to be achieved when the user is actually further from the base station. So, that's one area.
Secondly, enhancements in mobility, reducing delays in connecting to new cells as you move through the network, and also reducing delays in activating carriers to match the bursty traffic patterns of real-life data traffic, because real life data traffic is not just at a constant data rate. It varies depending on what you're doing. So, this will all help to improve the experience of the network operators’ customers.
Thirdly, another very important area, network automation. There's no getting away from the fact that 5G networks are complex to set up and optimize. That's a collateral of the immense flexibility that's built into 5G.
So, with 5G-Advanced, we will see enhancements increasing the possibilities for operators to use AI and ML, artificial intelligence and machine learning, to take care of some of that optimization, especially in areas like network energy minimization, load balancing between cells, between base stations, and optimization of mobility. So, plenty there for CSPs to get excited about.
MH: There's a lot of attention being put on artificial intelligence and machine learning when it comes to 5G-Advanced, and not just from the traffic shaping nature of it, but also the enhanced security that comes with it too. This is a ring fence around a ring fence around a ring fence that we never saw in 4G.
MB: Yeah, indeed. Security is absolutely critical to everything that we do, and increasingly so. We have to ensure that security of data that's transmitted, security and privacy of user information is protected.
Also, that the network is fundamentally reliable and that the operators and the end users can have complete confidence that when they're using a 5G-Advanced network, they know that data's going to be secure.
MH: Beyond data communications and security, I'm fascinated to learn that some of the new applications that we'll be able to get out of 5G-Advanced sort of put us right up against the GPS systems of the world. Are we putting satellites out of business here?
MB: That's a very good question. Indeed, I would see it not a question of putting satellites out of business, but I think part of the beauty of 5G-Advanced is that its capabilities are not just limited to simple data transmission. And of course, when it comes to positioning, we're all used to the idea that we can use GNSS, global navigation satellite systems, to identify our location for us when we're outdoors.
But we are also familiar with the fact that when you're indoors, the satellite signals can't be received. Quite apart from the fact actually that being dependent on satellite signals outdoors means also being subject to the vulnerabilities of satellites.
So, there is actually a real need for a super accurate level of location service that is consistent and reliable, both indoors and outdoors. And for that in 5G-Advanced, we're looking to introduce actually a technique that comes from the satellite world, an advanced technique called carrier phase positioning.
Now that's what's behind the ability to identify your location outdoors with an accuracy of a few centimeters, and 5G-Advanced will be the first time that that will actually have been deployed in a ground based mobile communication system, enabling you to take advantage of it when you are indoors as well.
So that's very exciting. It'll bring new opportunities for industrial automation, production halls, warehouses, airports, and the like that could do so much more automation if they have that level of location accuracy.
And I mean, while we're on the subject of satellites, another area that satellites are very useful for is timing. We have used many applications today where satellites provide us with accurate timing sources, but that's actually not enough. If you look for example, in the financial industry for automated stock trading, you actually need not one but two sources of very accurate timing. And that means you need a timing source that can be conveyed to the end user with the sort of microsecond or even sub-microsecond accuracy, wherever the user happens to be.
And that's another thing that 5G-Advanced will be able to provide you with, the distribution of absolute time information with the required levels of accuracy for those financial transactions, for the control of electrical power grids of the rapid switching, for example, to give you a completely different area.
So very critical that these non-data applications are built into the network, and to tie that together, I think the real beauty of that is that with 5G-Advanced, it can all be provided with a single network. You won't have to deploy additional beacons or multiple networks, but a single 5G-Advanced network can handle your data communication, your location requirements, your timing service. Can support users on the ground as well as users in the air with what are often known as, in common parlance as drones or in the jargon, we tend to call them UAVs or uncrewed aerial vehicles. All of that supported with one network.
MH: Well, with that in mind, and despite the fact that we really opened this conversation with more of a consumer-oriented experience of extended reality, etc. It strikes me that 5G-Advanced is also an enterprise upgrade to the wireless standard.
MB: Well, we have already talked about, of course, benefits to consumers in the XR domain. We've talked about the advantages for communication service providers in areas like coverage, capacity, mobility and automation. And then just now, the benefits of new applications like positioning and timing that are perhaps more immediately useful on the enterprise side.
So, the benefits of 5G-Advanced are diverse and they will be visible to consumers, CSPs, and enterprise and industrial users alike.
MH: So, if we're not expecting 3GPP release 18 until 2024, what must CSPs do to prepare industry for this Fourth Industrial Revolution?
MB: Well, the first step would be to deploy those features from releases 16 and 17 of 5G that we spoke about at the beginning, the URLLC features for ultra-reliable low latency communications, that provide the foundation of, for example, six-nines reliability and millisecond latency that are critical for automated manufacturing.
And then the additional features of 5G-Advanced, like the super accurate positioning and distribution of absolute timing can be added on top, creating a complete suite of features for the Fourth Industrial Revolution.
MH: Now, if it's been said that there's no going green without digital, what role does 5G-Advanced play in reducing carbon emissions?
MB: The first thing I would say is that 5G is already the most efficient mobile communication network ever deployed when comparing like for like data rates and capacity. And since the launch of 5G, we've already done a lot of work on the device side particularly to improve energy consumption and increase battery longevity further, both in standby mode and when active.
Now, one of the first things we're going to do next in 5G-Advanced is to look on the network side to create, first of all, a model for the energy consumption on the network side, to understand better where the energy is used and how new features will impact the network energy consumption.
We'll then be able to use that model to ensure that all the new features in 5G-Advanced that we design, use solutions that take network energy saving into account. And we'll be able to look at how we use different dimensions of operation, like the frequency domain, the time domain, the use of multiple antennas and the level of the output power to minimize the power consumption in any given scenario.
So of course, it's critical that we minimize the direct energy consumption of the networks, and we will be doing that in 5G-Advanced. Not forgetting of course, that the very existence of the advanced capabilities of these networks brings huge reductions in carbon emissions, by things like reducing travel and other sources of carbon emissions. And in 5G-Advanced, we will then see minimization of energy use being at the heart of all the new features.
MH: And I can imagine a lot of that is going to be AI driven.
MB: Absolutely. AI is one of the ways in which we can optimize the huge number of different parameters that we have in the network to ensure that in any given scenario, we are selecting the features and the modes of operation that help us to minimize our energy consumption.
MH: And as we talk about all of this 5G-Advanced and release 18 which is due out in a couple of years or so, I'm amazed we're already even talking about 6G. Why?
MB: A very good question, and 6G, first of all is still some way off. We wouldn't expect 6G networks to be deployed until the end of the present decade. But it is possible nonetheless to see 5G-Advanced as paving the way to 6G in a few areas.
And maybe just to continue on one of those areas, the AI and ML that you just mentioned. AI and ML will not only provide a benefit in network optimization, network automation, but it will also increasingly have the potential to play a role in the air interface itself.
So, during the 5G-Advanced era, we are going to be using 5G as a platform to study how AI and ML can improve the KPIs of the radio transmissions themselves. Now the hypothesis there is that AI and ML can enable new modes of operation to be developed on the fly to optimize communications in the prevailing scenario, which we might not have been able to fully predict beforehand.
So, example areas could be feedback of channel state information, for example, to facilitate multi antenna transmission and link adaptation. It could be optimization of beam management for the fundamental beam forming that's at the heart of the 5G air interface. It could be to facilitate further improvements in positioning. So, in each one of these areas, we'll be looking at AI and ML to see what these new techniques can contribute.
And we'll also have to consider very carefully how we would then go about setting performance requirements and carrying out conformance testing against those requirements in an AI/ML world, which is something we've never done before.
And we need to ensure that when AI and ML are introduced, they bring gains in performance rather than losses, even when the test conditions that we may be testing under are different from the scenarios in which a device might find itself in the real world.
So, we hope to find out during the 5G-Advanced era, how much improvement may be possible using AI and ML, and then in the future to create a whole new paradigm for the way that 6G will be designed. So, 6G would then be the first opportunity to design a new system where AI and ML air interface optimization are taken into account in the fundamental design from the very beginning, where we don't have that opportunity today because we have already the fundamental 5G system which is designed the way it is, and we have to maintain backward compatibility throughout 5G-Advanced.
But it does give us a platform to play with, a platform to investigate these new techniques, which will no doubt play a much stronger role when we come to 6G.
And maybe if I just mention another area as well which is of fundamental interest again to the CSPs, just to underline the relevance of 5G-Advanced to the CSPs. The flexibility of use of spectrum, and much of the spectrum that operators are using for 5G today is TDD, time division duplex spectrum.
And by regulation that typically has a fixed pattern of uplink and downlink time slots. And as we see variation in traffic patterns with new applications coming along, we see changes in the uplink/downlink traffic ratios, and then a fixed pattern of time slots starts to look much too constraining.
So, we're also starting to look in 5G-Advanced at how we can manage the uplink and downlink time slots more dynamically. How we can handle the interference that's generated when we do use the time slots more dynamically.
And in 6G, I'm sure we will need a high degree of flexibility in that area. So again, the investigations, the studies and the developments that we make during 5G-Advanced will be paving the way for the fundamental spectrum usage in 6G.
What we're seeing with 5G-Advanced is that we are now taking full advantage of that flexibility that was built in, and we are adding into 5G all the new features that it's capable of. So features that can improve the experience for the user, features that can enable 5G to bring new value to industry and enterprise, and features that can improve the operation of operators’ networks. And I think all of those are the full 5G vision that we set out with when we started down this journey of 5G, and we expect that to be fulfilled in 5G-Advanced.