This series examines the history, governance and radio spectrum technology that led to 4G and 5G today.
The first post in this series looked at the governance of the radio waves and how that opened the door to analogue telecommunications technologies. The second post examined the digital wireless technologies that built on those analogue predecessors — 2G to 4G.
Today’s post examines the latest ‘G’ to join the club. What’s 5G when compared to 4G and its predecessors?
5G is about speed and enabling IoT
5G is really about a significant jump in speed and cost. If 4G represented the deployment of networks that were primarily about the Internet Protocol (IP) instead of traditional mobile phone lines, then 5G represents the deployment of gigabit-capable IP services, and deployment of the Internet of Things (IoT). The 5G model has to encompass millions of online devices (IoT) sending typically low levels of data, alongside millions of online people using high-speed Internet to stream videos and play games.
So the next question is: ‘How different is 5G?’
Read more: Gimme a slice!
In principle, when 5G was first mooted, it was going to be radically different. This isn’t unlike 4G, which was supposed to be a quantum-leap above 3G, but actually turned out to be 3G with Voice over IP (VoIP) and slightly reduced hiss.
5G is also about directionality — sending beams
Manufacturers have proposed that 5G be designed to have ‘beam forming’ directed radio (where the base station uses complex digitally steered antenna, which shape the radio signal into a tight beam) to reduce interference. This is costly technology, but at the scale of deployment of a national mobile provider, costs could be driven down. By using directed radio signals, the amount of ‘noise’ (which encoded signals have to live in) can be massively reduced. This has big upsides; if you are using 4G as the sole occupant of a given cell (an allocated area of coverage) and then somebody walks into the cell and starts sharing your bandwidth, the reduction in your own speed is quite remarkable. 4G branded itself with ‘up to’ language (as in ‘up to 100Mbps’) but the reality would usually be something more like 20Mbps, with occasional peaks of 100.
5G offers the promise of more sustained high-speed use, because the beam-forming means you received stronger directed signals to your device, without interfering with another user. It’s not as much like a ‘cell’ in some ways, but it still depends on a cellular model.
Read more: The impact of 5G on IP transport networks
Fitting more data in less space
5G is about higher frequencies, but it can also be about more intelligent use (more complex encoding) inside older lower frequencies, which travel further. This is incredibly useful! However, the use of those frequencies depends on not sharing those frequencies with other encodings.
Higher frequencies mean more transitions in the radio wave, so there are more possibilities to encode digital data in any given unit of time. Simply moving into a higher frequency itself improves bandwidth, but adding smarter encoding (like being able to say more, or correct for more errors, or say more in less wide channels of radio space) means more can be said in the same old space. This is important, because as the world moved closer to mass adoption of 4G, it moved closer to the congestion of everyone online, all the time, in the same segments of the radio space.
Sometimes this also means competition for the same space. 5G is going to come with its own set of issues because as more services are pushed into the radio spectrum, the more likely it is that there will be points of collision. Already, American aviation authorities have expressed concern that 5G might encroach on the same bands used by altimeters.
Read more: Reducing the complexity of 5G networks using Segment Routing IPv6
5G means smaller cells, so more radio masts
In my economy (Australia) for the last 18 months, every second lamp post and power pole near where I live has been plastered with signs informing me of an application to install a 5G radio base station on this pole.
Installation of radio infrastructure has sometimes caused dissent, partly because the radio masts were typically very ugly, and had to be huge (to see as many people as possible).
But 5G base stations are different. Firstly, they are being mounted on what is sometimes called ‘street furniture’ which already exists; they are not new towers. And secondly, they’re tiny. They look like small white boxes, probably not much bigger than a suitcase. This is partly because things just tend to get smaller over time in the electronic communications world. But it’s also about the frequency issue in 5G. The higher frequencies travel less far for a given radio strength, because the high frequencies are blocked by buildings and other things like rain and trees.
The solution for 5G has been to deploy ‘micro’ cells, which fill in the gaps between the existing towers. 5G is using older frequencies more smartly, but also backfilling the radio spectrum with newer frequency allocations. They need more antenna to reach us all, but they send out signals that are inherently less ‘strong’ in propagation.
5G can use Wi-Fi and that’s a bit of a problem
5G was designed with an eye on the Wi-Fi ‘unregulated’ space (known as the 802.x series of standards) that powers hotspot services everywhere from your home through to trains, planes and the local shops.
Almost every computer is now built with Wi-Fi baked in, and the channels for Wi-Fi are well established. But some providers and 5G architects are proposing ‘repurposing’ segments of the public unregulated Wi-Fi spaces for 5G in a cooperative model.
This would be ‘cooperative’ to their benefit, and there is pushback from the consumer advocacy movement. Governance of the radio spectrum has to include segments of licensing that allow ordinary people to do things like ham radio and Citizen’s Band (CB) radio but also do things like running their own private data networks.
Cellphone carriers would love to monetize this. There is the argument that it ‘is better use of the spectrum’ but this seems like a slippery slope argument that leads to worse and worse Wi-Fi at home, for their benefit.
What’s next? Will 10G arrive someday?
This ‘G’ model isn’t going away. In some ways 5G has turned out to be ‘4G with extra bits’ just as 4G was ‘3G but better’. 3G was actually pretty novel compared to 2G, and all the Gs are digital unlike the 0G and 1G analogue phones that preceded them.
So, does the technology get smarter? Yes! Chips get faster, there can be more computing in the same space, or for less energy (longer battery life!). It becomes possible to handle more and more complex encoding schemes that reduce error and make radio channels capable of being even narrower. The technology may move into near-visible spectrum (modulating light is what drives fibre-optic communications but can work in principle free-air without a fibre wave-guide). It seems inevitable that there will be progress from 5G to 6G, 7G and 8G in time.
As with the ‘2G turn-off’, it’s coming to the time when 3G gets turned off as well. The 2G turn-off has now been completed in many economies, but actual use of 2G encoding (EDGE, GPRS) continues as the fallback from 4G, and in due course 3G bandwidth will be repurposed for 5G and future use. There are already places that are reusing older analogue TV channels (which go through brick and concrete quite well) already.
It’s unlikely the world is heading to less data. Therefore, it’s probably heading for a revision in the data carriage of radio waves. Sometimes less (frequency) is more (data) so improving the use of what is there is important. The industry and governance models here really are seeking the same thing — better use of the radio waves that the world is bathed in, from sunrise to moonrise, natural and artificial.
The views expressed by the authors of this blog are their own and do not necessarily reflect the views of APNIC. Please note a Code of Conduct applies to this blog.
While there are some 5G bands that overlap with spectrum that can be used by Wi-Fi (nr-U bands), I haven’t seen equipment, handset support, or carrier demand for them. 4G also supports the same access to ISM spectrum (LTE-U) but it’s pretty lightly used.