Getting the measure of LTE

While operators and network vendors tend to make a name for themselves through headline grabbing deployments in the LTE sector, there are companies working in other areas on the new technology that command less of the limelight but nonetheless play a vital role in a successful LTE network roll out. Test and Measurement is one such area, and JDSU one such company. And, as Per Kangru, head of Business Development for LTE at JDSU, tells Telecoms.com, he is more than aware that testing isn’t usually at the forefront of everyone’s mind when it comes to roll-outs.

Indeed, Tommy Ljunggren of TeliaSonera is famous for exhorting operators to forgo rigorous pre-launch testing altogether telling them to, “stop doing trials and launch your networks.” Of course, this was back when TeliaSonera was a lone LTE pioneer, anxious to kickstart the LTE ecosystem. But did his views, and TeliaSonera’s decision to run its testing programme in tandem with commercial launch of the service, set a dangerous precedent?

“It depends on who the upgrading operator is, what the market is and what time frame they are working in, “says Kangru, diplomatically. “Compared to the largest international telcos, TeliaSonera is a relatively small operator. So, from that perspective, it was a very different situation than if such an approach had been taken by a carrier like Deutsche Telekom, which has a very different position in the market.”

The implication is that, for bigger players, spending time on test and measurement before launch is essential due to, as Kangru describes it, “a competitive need for premium quality”.

Not that test and measurement is concerned only with ensuring quality, Kangru says. Working with a T&M vendor in the early stages of deployment can help carriers to improve their cost management, he suggests. “We often have an extremely productive dialogue with operators, even if they are not in that much of a rush to deploy LTE,” he says. “It can help operators work out which are the top three suppliers that they could use, and then use the results from those trials to push the vendor pricing down. Operators can save a significant proportion of their CAPEX—they can basically play the suppliers against each other”.

Some LTE rollouts have been criticised specifically for the poor performance of the handover between 3G and LTE and Kangru says that in these cases operators haven’t done sufficient work on their network. “In most cases you are able to adjust network configurations to deal with problems—if you are aware of them. Operators can make a network configuration change so you can [first] minimise the problem in the lab,” he says.

Though he won’t mention any names, Kangru also has strong views on the operators who did not properly test when rolling out their HSPA upgrades, explaining that there’s always a difference between the lab and the real world. “The problem was that when some of the networks made it out under real conditions, some networks behaved extremely poorly and some networks behaved very well. The operators didn’t have the tools to help them qualify the real behaviour of the base stations—they just looked at the download data”.

The implicit advice here is that that whether upgrading to HSPA or making the full jump to LTE, it pays to make a vendor choice based on a solid understanding of how the network will behave in the real world. “You will start to see very significant performance differentiation between the vendors when they have loaded networks and how they deal with that load,” Kangru says.

JDSU works with many infrastructure vendors, giving it a decent perspective on their strengths, opportunities and weaknesses. Kangru offers a forthright take on the prospects of Samsung, which revealed at the recent LTE World Summit in Amsterdam that it intends to enter the crowded European network telecoms vendor market. “Samsung will need to do something special. You need to do something fantastic… something that is not available from the other players. It may have to offer a deal or package that is too good to deny in order to establish a beachhead in the industry.”

JDSU’s work also allows for insight into spectrum usage and Kangru expresses surprise that European operators have not looked more closely at TDD licensed spectrum. “I’m surprised that the TDD licensed spectrum isn’t getting more attention. There are a range of companies that are buying up spectrum very, very, cheap,” he says, specifically pointing out the acquisition of TDD spectrum in Germany and Denmark. Most networks are being deployed as FDD and Kangru believes the explanation is very much technical.

“Most people don’t have the background in TDD so they don’t see that it’s quite easily deployable and they haven’t really seen how well the LTE specification has been designed to integrate TDD. If you look at the publicly announced plans for the LTE chipsets you can see that almost all of them have support for TDD. So you’ll actually find most phones being dual mode – generation two and three chipsets will do that.”

Kangru also points out that using TDD is a good fit for how people use data services—with more downloading than uploading.  “It’s a good fit for asymmetrical. Normally in mobile data people download more than they upload so with TDD you can have an asymmetrical ratio—80 per cent downlink and 20 per cent uplink as an example, whereas with FDD you only get 50/50. So you can use the spectrum more efficiently.”

So while test and measurement expertise may not be high profile, Kangru makes a strong case that working with JDSU can make for an operator looking at LTE, be it with pre roll-out planning, post launch fixes and general technical advice. If this helps deliver a faster, more reliable network, everyone, be it shareholders or customers, will end up better off.