What Advanced Applications Will We See On The LTE Network?

Verizon Wireless announced that it is updating its specifications for LTE devices that will run on the network. These new specifications are aimed at helping developers to design products and services that will run on the operators' LTE network, which is scheduled to launch in 2010 in 25 to 30 US markets.

Verizon Wireless said the updated specs will address network access, SMS requirements and data retry test plans. In addition, new information about lab and signaling conformance, open development device approval will be included. The new specs will be outlined in a webcast on the 20th of January.

What will be interesting to see after the webcast is what new applications and business models the application developers come up with. VZW wants the LTE network to expand existing types of applications while opening up entirely new classes of applications at the same time. One example that has been pointed out by VZW is the possibility of appliances such as refrigerators being fitted with wireless monitoring devices. A missing part, for instance, could be pinpointed via a wireless link, cutting service costs. Innovation in advanced video and gaming services are also envisioned by VZW.

It will be interesting to see if other operators deploying LTE at the moment push as hard for new devices and services for LTE as VZW has. TeliaSonera for instance, who have already launched their LTE network, has an agreement with Samsung who is providing it with a USB dongle. There does not seem to be a plan for anywhere near the same level of activity that we have seen with VZW when it comes to launching new devices and apps. So do operators need to go to all the trouble the VZW is?

Alcatel-Lucent and Ericsson are the "founding participants" at VZW's LTE Innovation Center in Boston, MA. A group of venture capital firms are also participating in the core working group at the center. In addition to Alcatel-Lucent and Ericsson, other participants in the core working group include Charles River Ventures, Northbridge Venture Partners, Norwest Venture Partners, New Venture Partners, and Redpoint Ventures.

An interview with Tony Malone, CTO of VZW can be found on the LTE World Series You Tube channel. Tony spoke at the LTE Americas conference in Dallas in November 09.

Is the LTE Performance Gain Simply Due to Wider Frequency Bands?

There have been various LTE trials going ahead in 2009 and of course the surprise LTE network launch by TeliaSonera who have estimated LTE user data rates to be in the range of 20-80 Mbps. Similar figures have been promoted in other operator trials.

Quite often operators and system vendors refer to the achievable peak rates, ranging from 100 Mbps to as high as 250 Mbps. However, in cases where LTE is deployed in high density, metropolitian areas, these peak data rates are unlikely to be achieved.

Omnitele has just announced that it expects actual data rates to be a lot less than the figues above. Through analysing LTE performance in technical studies and simulations using Omnitele’s state-of-the-art network planning tool analysis on the performance gain of LTE compared to HSPA technology comes mainly from the wider frequency band (up to 20MHz compared to 5 MHz for UMTS). Switching from CDMA to OFDM also has an effect, as does MIMO according to Omnitele. However, significant expectations being put on the performance of MIMO and yet the most critical element of performance which remains under the control of the designer is the antenna, The 3GPP is still proposing how to define requirements for MIMO antennas and it is a pretty complex topic with apparently little consensus developing so far.

Actual LTE user data rates are highly dependent on radio conditions and number of users sharing network resources. Most of the first commercial LTE deployments are said to utilise 20MHz bandwidth and 2x2 MIMO antenna schemes. When breaking down the performance of LTE features in different channel conditions and simulating them in a realistic metropolitan network environment, Omnitele estimated the achievable average LTE user data rates to be in the range of 15-25 Mbps per 20MHz frequency bandwidth.

LTE outperforms the current baseline HSDPA in terms of data rates by a factor of ten and HSPA+ technologies by a factor of 3-4. But it will take more than improved radio performance to really get the best out of LTE.

Is Network Sharing and Joint Ventures the Best Model for LTE Deployment?

With the joint venture announced between Telenor and Tele2 in Sweden, is it just too expensive for operators to be building their own LTE networks?

Operators such as Orange and T-Mobile in the UK are already looking to merge their 3G networks, at great cost. £600m-£800m is the estimated cost that will be taken up by decommissioning redundant duplicate radio network infrastructure, as well as reducing the number of retail outlets and combining the customer service centres and general administration functions. T-Mobile will contribute the 50% share of their joint radio network with Hutchinson 3G to the pot, (who incidentally already use Orange’s 2G network for fill-in coverage). Assuming T-Mobile and 3 put both their radio networks into the joint venture, you’ll end up with the interesting situation of 3 using a joint 3G network shared with Orange and T-Mobile, and a GSM network operated by Orange and T-Mobile.

So what does a joint venture between two operators look like? The Telenor and Tele2 merger seems a lot more simple. They have created Net4Mobility, a company that is a product of the joint venture that will build and manage the joint network for the two operators, have a competely new infrastructure (radio, backhaul, core, OSS etc). The joint venture will be 50/50 between the two operators. Net4Mobilty will be using its own 2.6GHz spectrum and will also use both Telenor's and Tele2's 900MHz spectrum.

Sharing spectrum and network infrastructure massively reduces the CAPEX and OPEX when compared to the investment that would need to made if the two operators deployed LTE seperately. But is this also a move to compete on customer experience and product differentiation? The Swedish market is one of the most competitive for mobile broadband and perhaps the operators have decided that they simply cannot continue to compete on who has the cheapest flat rate plan for data.