Why is 5G mobile technology such a mystery?
March 20, 2019
The debate around 5G — an embryonic technology that's seemingly poised to replace 4G and meet our ever greater "demands" for mobile data for streaming films, navigating self-driving cars, or getting rich on blockchain — is as much a war about patents as it is about business, security, and, well… thinking that everything that the Chinese do has just got to be bad.
China's a dictatorship, right? So, they're going to want to snoop and surveil, and probably snatch your first born, too.
"Democracies," on the other hand, like the USA, Germany and UK, are as pure as Columbian snow. …We've never heard of the NSA or GCHQ, either.
Read more: Germany won't ban Huawei from 5G auction
But the list of countries seeking to ban Huawei — a Chinese company — from future 5G networks is legion.
The USA tops that list, but the fact that an American firm called Qualcomm has its own star-spangled ambitions for 5G will have nothing to do with that.
The Americans are followed by Australia, New Zealand, a currently no-so-great Britain, and Canada.
All five nations are part of the "Five Eyes" spy alliance, while others, further down the list of those either considering or threatening a Huawei ban, include countries like Denmark and Germany, who feature on an expanded lists of "Nine Eyes" and "Fourteen Eyes," respectively.
They've all got it in for Huawei, and yet none of them can quite pull the trigger.
Huawei must be packing something special
The main players cited in 5G tend to be Qualcomm, Nokia, Ericsson, and Huawei.
Each has their technological pros-and-cons — they are competitors, proposing alternative ideas, a bit like the VHS-Betamax wars in the 1970s and 80s, or DVDs vs. Blu-ray, or streaming vs 180-gram vinyl today.
As a result, 5G technology is all "still up in the air," as Muriel Médard, a professor in Electrical Engineering and Computer Science at the Massachusetts Institute of Technology (MIT) in the USA, has described it.
And that may also explain why Huawei remains a contender, despite various concerns that it's just a front for the Chinese government and / or military.
"Huawei has a good number of patents in 5G technology, patents that other companies don't have," says Torsten Gerpott, a professor in Technology and Operations Management at Duisburg University, Germany.
"I wouldn't necessarily say that Huawei has an advantage over other companies, but they are certainly the 'leading edge' with Nokia and Ericsson," Gerpott says, "and as they're built into existing 4G networks, it will be hard, although not impossible, to substitute them technologically."
After some light prodding, Gerpott goes on to list a few of Huawei's other strengths.
He mentions CDMA technology, core network IP, and frequency-division multiplexing — all jargon to you and me, but vital for the next generation of mobile telecommunications.
Read more: Mobile World Congress is about conversations, not devices
A lot of this technology aims to maximize what can be done on a single network.
Frequency-division multiplexing, for instance, is a method to split "bandwidth" — a cable, optical fiber, or air waves — so that different streams of data, whether that's phone calls, radio shows, or cash transactions, can flow without overlapping or interfering with each other.
Similarly, core network IP and CDMA technology are concerned with enabling various types of data to flow over a network, whether that's voice, data, video, games, GPS tracking, or navigation for cars.
In 5G's case, Huawei says this may be the data of up to a million people in a single square kilometer.
Silence of the codes
How any of this will work depends largely on how a network is programmed and the efficiency of its code.
You know how some computers boot up quicker than others, some browsers search faster than others, and some webpages load like slugs. Well, that's all down to the efficiency of their code… like a well-oiled car, or a well-doped athlete.
And when it comes to 5G, there seems to be a battle between Huawei's "polar code" technology and Qualcomm's standard of "LDPC" (low-density parity check).
Gerpott's view is that the two are merely alternative approaches to the same sort of thing. And both are supported by the 3rd Generation Partnership Project (3GPP), a group which is overseeing the standardization of 5G technologies. So they are equals in a sense.
Roughly speaking, polar code and LDPC are descendants of the "turbo code" which enabled 3G networks.
The new code, however, will have to facilitate data transfer speeds of up to 100 times faster than our current standard of 4G/LTE… And by the way, 4G is just 10 times faster than the previous generation, 3G.
But that doesn't really explain a lot to the average data abuser, the kind of person who makes a video call when they could just send a text.
So I contacted the so-called godfather of polar code, Dr. Erdal Arikan of Bilkent University in Turkey, to find out more.
DW: Why is Polar Code so significant in the 5G story, and how does it compare to Qualcomm's LDPG?
Arikan's response, while not swift, was succinct:
EA: I cannot comment on any of the questions you are asking simply because I do not have factual information regarding such issues. I hope you will understand.
Alas, to be honest, I did not understand. So I sent this:
DW: I thank you for your kind response. Just to be clear, though, would you not be in a position to explain how Polar Code works, explain its defining features, even without reference to anything else or any other company?
EA: No, I do not wish to talk about my own work. You may ask Prof. Johannes Huber for an expert opinion.
Hmmm…
DW: Sorry... but which Johannes Huber do you mean exactly? Where is this Prof. Huber based? I'm getting results for a member of Germany's Alternative für Deutschland party, and I'm pretty sure that's not whom you mean.
I also found a gynecologist Huber, and briefly wondered whether he was some kind of Renaissance man, who dabbled in code on the side.
Until Arikan fired back:
EA: https://www.eei.tf.fau.de/person/johannes-huber/
Ah…
So I contacted Professor Huber at Friedrich-Alexander University in Erlangen-Nürnberg, and, at time of writing, await his reply.
The same goes for Prof. Médard at MIT (traveling), and Prof. Christel Baier at Germany's 5G Lab in Dresden (silent).
A gigabit feast
No doubt there will be occasion to speak further on 5G as the technology is a feast moving at gigabite speeds, working towards, what Gerpott calls, a "family of standards."
Perhaps the guardedness among some of the players can be explained by the fact that their technologies are all so very similar — whether it's their development of Massive MIMO (multiple input multiple output), a technology that should ultimately allow the simultaneous flow of many data signals over a single radio channel.
Then there's their research into millimeter waves, which can support broader bandwidths and a lot more data, moving much, much faster.
Millimeter waves may also allow for "fog computing" — a form of peer-to-peer, mesh networks in the cloud — perhaps good for emergency situations — but that will take a heap of clean code to get right, according to Médard in an interview for MIT News.
And the next 3GPP standard (Release 16) is not due until late 2019.
So, for now, we'll just have to put up with good ol' 4G, and to be frank, it'll probably do for most of us.