As the first operator to put a 4G mobile network based on the Long Term Evolution (LTE) protocol into action in the UK, EE will have to look abroad to take lessons from other deployments.
4G networks around the world have to date started as isolated pockets of mobile broadband that act as supersized Internet hotspots primarily in urban areas. There are both technical and commercial reasons for this approach. Offering small pockets of coverage helps eke the most out of limited radio spectrum without the waste of overlapping cells, which helps to save on initial investment costs before a subscriber base has accumulated.
The pockets-of-deployment tactic also reduces the technical risk, albeit with knock-on effects. In contrast to older networks, where adjacent cells try to use different frequencies within the network’s allotted bands, LTE tries to maximise radio usage by using the same frequencies across all cells. This increases the chances for interference, so handsets and basestations need to adjust how they work together. This has yet to be tried on a large scale and the current approach is to avoid 4G cells overlapping.
“As you move, the call has to hand down to 3G and then come back to LTE when you are in range of another 4G cell,” says Jonathan Borrill, director of marketing at communications-test specialist Anritsu. “3G had its own interference issues when it first started. Now we have to solve them for LTE.”
The handover from 4G to 3G for voice calls raises problems as the speech is carried on an Internet-style packet-switched network on LTE networks; but this turns into a circuit-switched system, more akin to traditional telephony on 3G. The issue for young 4G networks is that this transition can happen frequently and cause issues for handsets that are not designed to cope.
“One lesson learned from the US deployments is that the existing certification scheme for handsets was not enough. Operators have found that there are so many issues to look at, such as roaming, handovers and that no one has a standard network,” says Anritsu's Borrill.
The key reason why the EE network is arriving ahead of the UK’s first dedicated 4G spectrum auction is that EE is borrowing radio spectrum that was licensed by EE’s component groups Orange and T-Mobile in the 3G auctions more than ten years ago. Where the existing cells from Orange and T-Mobile overlap, EE has the opportunity to reuse some of the radio capacity for the newer, higher-data-rate LTE protocol. However, the EE plan has a cost attached to it in terms of coverage.
And the EE band at 1800MHz will suffer signal degradation in buildings, not unlike the problems many see with 3G, according to broadband information bureau Thinkbroadband: “This degradation is why the 800MHz band in the Ofcom spectrum auction is likely to be the most popular one.”
In the US, where 4G deployments have attracted several million customers so far, network operators have been able to use the lower-frequency bands in the 700MHz range, freed up by the demise of analogue-TV. To improve coverage, EE has the option of relaying data to handsets with multiple-input, multiple-output (MIMO) antennas. Several EE-approved devices made by Huawei offer MIMO, which lets the handset look at signals that have taken different paths to reach it – typically reflected off buildings.
But, says Jonathan Borrill of Anritsu, the MIMO approach has its own issues that will affect network planning and deployment. “MIMO makes it quite difficult for a planner to predict what kind of coverage they might get. The MIMO algorithms almost doubles the data rate but because it uses non-line-of-sight signals, is harder to model. It will take time to get the practical experience that network planners need.”