“Radio access alone is not enough for us to establish a really good market,” Ewaldsson says.
With a few years to go before the dawning of the 5G era, Ericsson has developed a new radio system that is small enough to fit into a suitcase. And according to the marketing material, it can be set up in a matter of minutes.
Another newsworthy radio innovation is a technology that will compensate for the fact that 5G has to operate on a higher frequency band (with worse coverage) that its predecessors. The innovation, which is called “beam tracking”, allows our telephones to gain speedy access via the most suitable antenna at any given time: changes can take place several times within a single millisecond. The antennas themselves are new, and are more advanced than those built previously.
Anyone who follows the company closely will already be familiar with the apparent advantages of 5G: network speeds and response times that – according to Ericsson at least – supposedly compete with those of fiber optics.
“Fixed and mobile networks are converging,” Ewaldsson says. “In fact, in 5G technology there will be no distinction between fixed and mobile. This means that the dream of ‘the last mile’ over wireless will now become a reality.”
‘The last wireless kilometer’ that he is talking about is the possibility to offer broadband-like conditions to places fiber cable has not yet reached. One example is in Boliden’s mine in Kankberg, outside Skellefteå, where 5G testing has been going on for a year or so.
But for Ulf Ewaldsson, all this constitutes only one part of Ericsson’s definition of 5G. Another part – one that is at least equally significant – is the cloud.
The telephone has been connected to the internet for a long time, as have the computer and the TV. Now it’s time for all the other gadgets to take a similar step.
“This means you can have a virtual infrastructure in the actual network,” he says. “A processer that doesn’t know which batch is being carried on it.”
With a solution like this, special hardware for a particular purpose becomes redundant. We are accustomed to thinking of computers as units made up of processors, memory and hard discs connected by electricity and neatly packaged in a chassis. But Ericsson’s idea (one that they certainly share with others) is to take the units apart and allow their components to communicate with one another instead, using light.
Jason Hoffman, Head of Product Area Cloud Systems at Ericsson, says: “The idea is that nothing needs to function together in the same box. The central processors, memory and hard discs become individual entities. We then connect them with a buss – a kind of cable – and create a single optical network.
“Most transport networks are optical, which is ideal if you want to make datacenters completely optical.”
Hoffman uses an example: a processor on a central circuit board can use a hard drive located 3km away, but the user remains unaffected by the geographical distance. By extension, this means that virtual applications and entire systems can be transferred to and run in the cloud.
“If you are completely optical, you can virtualize everything,” Hoffman says. “Absolutely everything.”
When we think of cloud infrastructure today, we often think of the big players: Google and Amazon, Microsoft and Apple. We don’t think of the tens of thousands of datacenters around the world that also deliver IT systems. Ericsson recognizes the opportunity that these represent. Because what happens if they can build an optical network that connects all of Ericsson’s millions of base stations, with tens of thousands of modernized datacenters and a public cloud?
“Then we can start building control systems,” Hoffman says. “We can start considering things in the same way as we consider the human cardiovascular system.”
An optical network, systems virtualization and a data cycle. To Ericsson, these are also 5G… not least because they have created the best possible conditions for the Internet of Things. The connected devices will not become smarter just because they are connected, or because they measure things with sensors. They need to deliver new insights too.
“You know, we have built infrastructure in almost 200 different countries,” Hoffman says. “But what is it used for today? It’s used for media consumption. It’s used for taking selfies. But here’s what we’re thinking: what would happen if we started sending data upstream from several trillion devices instead? And what would happen if we could take decisions based on that data, and change the behavior of the thing the data came from?”
For Ericsson and their competitors, the race is on right now: to become the IT supplier for the systems and platforms that will handle the data, to make the applications that will deliver insights about that data. The companies that in turn buy the IT systems do not need to invest in hardware; they can rather order it according to their needs: on demand and ‘à la carte’. Ewaldsson notes that there has been a marked cultural shift, with Ericsson no longer so closely tied to the operators.
“This is the ultimate kind of infrastructure. And as you see, it is independent of the operators. Some say that if things go badly for the operators, things will also go badly for Ericsson. That might be the case, but not necessarily. We are infrastructure suppliers in a system like this, and it’s up to the operators what they want to do with this infrastructure.”
“The transformation we face now is as big as the one we went through when the switches were digitalized in the 1970s. I can clearly see the huge investment being made in R&D in Sweden, as one of the key aspects of this process. I would like to address this powerful message to Swedish industry as a whole, as well as to universities and colleges: that this is what we want to invest in. We would really like to gain the support of talented, knowledgeable engineers, to take our offering in the direction I have just outlined.”