Ulf Ewaldsson: It has always been important for Ericsson with
a new radio technology that we are first on the market — it gives
us important experience 
                             
                             
The mobile industry has in its grasp the possibility of a single unifying radio technology that will deliver all wireless broadband services for years to come.
The single technology is LTE — long-term evolution — the fourth generation technology that is part of the GSM family, but has been adopted by operators that previously used CDMA rather than GSM for 2G and 3G.
“One radio technology: that is the ultimate goal for radio access,” says Ulf Ewaldsson, vice president and head of radio at Ericsson. “And it’s no secret that some of the first operators to adopt LTE are from the CDMA world.”
The implications of a single unifying technology are profound, not only the developed world but in emerging markets too, he says. Thanks to LTE, the smartphone will “bring the internet to the masses”.
But building a worldwide set of LTE networks is a huge technological and engineering challenge, he admits — though it’s a field in which Ericsson claims the lead. “We are the true leaders in LTE. We were there at the very early stage when the standards were set, and we have the most essential patents,” says Ewaldsson.
TeliaSonera launched the first commercial LTE service, using Ericsson technology, in December 2009. Before then Verizon Wireless had selected the company as an LTE supplier.
“This follows our tradition. We were the first with GSM, with a system in Germany, and we were the first with 3G, in Japan in 2000,” says Ewaldsson. “It has always been important for Ericsson with a new radio technology that we are first on the market — it gives us important experience.”
One of the benefits of that experience is knowing how to manage the launch of a technology, he notes. When 3G was being developed the “was an explosion of tweaks to the specification” which had to be incorporated and tested before the main code was stable.
“It’s a problem when people want all these features from day one. Developers end up doing too many things at once. With LTE it is extremely important to make sure of the basic functionality — features such as setting up and taking down connections, throughput, latency and radio performance. They are the most important, and that strategy has paid off. We in Ericsson are getting high volumes to market. The US market is recognising that we have high performance.” 
                           
                           


                           
LTE is taking off faster than previous technologies in the family, including both GSM and WCDMA 
                           

                           
Early to market 
                           
Being fast and first is a key part of Ericsson’s strategy, says Ewaldsson. “You have to be fast to bring the technology to market, so you have the most experience in the technology and you can deliver a stable, high quality system.”
And it has been quick: “In less than seven years we have brought the entire technology to market,” says Ewaldsson, who compares the speed with previous mobile radio technologies. In terms of adoption and subscriber numbers LTE is outstripping its predecessors, he says.
But it is complex. LTE has to live side by side with 2G and 3G technologies, and work with them. “LTE doesn’t include voice technology at the moment, so voice calls have to be made through a circuit-switched fallback using 2G or 3G.” That means handsets will have to run multiple technology, and the switchover “has to happen so quickly that the user does not notice. The technologies have to interwork.”
Voice over LTE — VoLTE — will be introduced in the near future, but that will also need to fallback seamlessly to 2G and 3G so that a customer making a call from a car or train does not lose the call between an LTE cell and a 2G/3G cell. “It’s very important that it’s seamless,” says Ewaldsson. VoLTE is already being tried out “in the US at the moment ... on several fronts”, he adds.
LTE is more than just a new radio technology: it is speeding up a series of rapid changes to the whole ecosystem that involves over-the-top players and providers of cloud services. It’s more than convergence, he notes: “Thanks to LTE, the computer industry is crashing into the telecoms industry, and it’s very clear that the US is going to be one of the most important markets for innovation, driving the ecosystem forward. The services from LTE are going to be launched first in the US. Apple and Google are strong drivers of this ecosystem.” 
                           
                           
Best user experience 
                           
That means that operators are being driven to ensure the best possible performance and deliver the best possible user experience with the new radio technology and with the new terminals that are being created.
“The network is going to be the key differentiator for operators,” says Ewaldsson. “Operators will need to supply the best possible performance with these new terminal products.”
But that means they will face some new challenges — including tying “all the network technologies together in terms of spectrum” — in other words, integrating the radio access network. Operators will need more and more spectrum in order to deliver the service their customers demand, and as LTE becomes the dominant technology spectrum that is currently used for older mobile standards will be reassigned — or refarmed, in the industry’s jargon.
Ericsson is already actively working on the next stage beyond LTE, which is LTE Advanced, being developed to cater for higher and higher bandwidth requirements, with more and more of the available spectrum using the same technology.
But spectrum does not come in handy chunks: not always, anyway. So there is a need to unite disparate spectrum together so it can be used for the same LTE exchange. “This is spectrum aggregation,” says Ewaldsson. “You put together different sections of spectrum up to a total of say 60 megahertz. It doesn’t have to be contiguous.”
Using the latest modulation techniques, 60 megahertz can carry up to a gigabit a second of data. “And that’s what will be required to make the capacity of these networks keep up with the traffic growth,” he adds. “Traffic is more than doubling every year.”
No one is suggesting that a smartphone or a tablet will be continuously receiving data at a gigabit a second, but if a peak rate such as this is available, the data can be transmitted when necessary with the spectrum that is available at the time. “Doing it this way is more efficient for the network and it means every user gets their work done faster,” says Ewaldsson. “Radio is an ever-changing environment but this technique allows you to send the maximum throughput to each terminal.”
Ericsson’s strategy of being first into the technology means that the company has the longest experience, can start work with operators first, and “and has the opportunity to work with the new terminals early and on the trials early”. 
                           
                           
Challenge of mobile data 
                           
This is critical, as “operators are trying to understand the challenge of mobile data that the smartphone is creating” and Ericsson is working with them from the start.
The smartphone revolution has already started, with the new terminals already starting to dominate the data business, and LTE will push that even faster, especially as the cost of new smart terminals is coming down.
LTE is also going to have an impact on the fixed broadband market, says Ewaldsson — not by replacing fibre to the home, but “but we see LTE as an extension of fibre networks, providing the last mile”, he says. Base stations will inevitably be connected to the network via fibre backhaul, and Ewaldsson sees LTE as an ideal technology to connect fixed broadband customers in rural areas to the network without having to lay a fibre down their trackway.
Australia’s carrier-neutral National Broadband Network issued Ericsson a contract to deliver LTE for just such an application in June 2011, for connections that would be too expensive to connect directly to fibre.
The NBN project will be using the TDD, or time division duplex, flavour of LTE, which uses the same frequency for uplink and downlink — unlike the more common FDD or frequency division duplex version. “We have been working on this technology just as long as FDD, though the FDD market started earlier,” says Ewaldsson. Which is better? “There are many different cases. It depends on traffic patterns.” In Germany for instance Vodafone is using the digital dividend 800 megahertz band to offer last mile connection in rural areas via the FDD flavour of LTE.
A consensus is emerging from the industry about which parts of the spectrum that will be used for FDD and which for TDD: “We’re seeing 2.3 gigahertz going to TDD, with 2.6 gigahertz for FDD,” he says. At the lower end of the spectrum, 700 megahertz is usually FDD, and “when 800/900 megahertz GSM spectrum is refarmed that will most likely be FDD”, Ewaldsson adds. 
                           
                           
Terminal choices 
                           
But he is wary about the fact that many different bits of spectrum that are being used for LTE around the world, especially in the early days of the technology when it is important to ensure that the market is standardised as much as possible. “We must make sure we don’t get too much fragmentation,” says Ewaldsson. “That means the industry has to make some good choices about terminals. There will be more and more terminal variants but if a terminal works on many parts of the spectrum that will drive costs up.”
That will become less of a problem as LTE matures, but it is key to avoid too much fragmentation in the early days, he repeats.
Mobile broadband is already a key technology enabling the world’s businesses and consumers. The benefits of LTE will allow huge improvements in performance, not just in the developed markets but in all parts of the world. GTB