The University of Surrey impressed this week with the launch of a 5G testbed at its shiny new 5G Innovation Centre (5GIC). It promises to put the U.K. towards the top of the class when it comes to developing the next generation of mobile technology; however, the lack of allocated millimetre wave (mmWave) spectrum may hamper its progress.
The 5GIC’s testbed is an impressive set-up.
It consists of three clusters, each comprised of one macro cell site operating LTE-Advanced TDD kit, augmented by around a dozen LTE-A TDD picocells. Within the walls of the facility are six LTE-A TDD small cells, six LTE-A FDD small cells, and six WiFi access points based on 802.11ac. The baseband units for all these sites are connected to two core networks.
So far, so 4G.
However, in its official blurb, the 5GIC insists its testbed is representative of an "ultra-dense environment in which to develop, test and prove three key 5G technology aspects: mobile broadband radio, evolved network infrastructure, and the Internet of Things (IoT)."
What’s more, as progress is made and the industry gets a clearer idea of what 5G will look like, the 5GIC said its testbed will "migrate to 5G, one cell cluster at a time". The upgrade is due to start in 2016, so that by 2018 it will boast end-to-end 5G infrastructure capable of delivering 10 Gbps of throughput per cell.
There is a potentially important part of the picture missing though: mmWave.
As far as the U.K. is concerned, Ofcom has suggested that spectrum in the 10 GHz, 32 GHz, 40 GHz, and 66 GHz bands would be appropriate to use for 5G. However, there won’t be a definitive answer on which mmWave frequencies will be globally harmonised for 5G until the ITU’s World Radiocommunication Conference in 2019 (WRC-19).
The 5GIC "does not want to invest a lot of resources in a lot of different frequency bands," said Rahim Tafazolli, director of the 5GIC, at a press conference on Tuesday.
"We will spend more resources once the [mmWave] frequency bands have been specified by the WRC-19," he said.
Allocating and releasing mmWave spectrum solves the problem of availability, but it does not address the challenge facing device makers, which is that the power required to process mmWave signals will drain a handset’s battery in the blink of an eye – if it doesn’t melt it first, of course.
For the 5GIC, addressing that problem involves reducing power consumption on the network, particularly signalling networks, while silently praying for a breakthrough in battery technology over the coming decade.
Nonetheless, "for millimetre wave, the number one issue is spectrum availability," said Wen Tong, vice president of Huawei’s Canada R&D centre.
He explained to reporters on the sidelines of the press conference that WRC-19 marks the official beginning of the mmWave allocation process, not the end.
"When that congress finishes it will identify a few pieces of potential bands for commercial use, and then, typically it will take another two-to-three years to release those bands," he said.
That poses a challenge for the likes of SK Telecom, which aims to launch a ‘pre-standard’ 5G service in time for the Seoul Winter Olympics in 2018.
In addition, Russia’s MegaFon plans to have a 5G trial network up and running during the 2018 football World Cup.
"I think it’s fair to say, 5G will probably come in two major releases, similar to LTE and LTE-Advanced," said Andy Sutton, principle network architect at EE, at Tuesday’s event.
"We will see 5G in sub-6-GHz spectrum, and then we will see 5G come in at the higher spectrum bands," he said.
Comments like that will doubtless have the marketing departments of every telco around the world salivating at the prospect of being in a position to effectively launch ‘diet’ 5G, followed by the ‘full-fat’ version a few years later.
Here’s hoping they come up with something a bit more inventive than 5G+, but I won’t hold my breath!
