By picoChip and Continuous Computing

LTE is coming. It’s almost impossible now to pick up an article on network infrastructure without reading about it and how companies are working on towards the shift.

The wireless networks in place today were primarily designed to support voice, a job that they actually do quite well and with plenty of capacity to spare. However, as we know, the majority of traffic on networks today isn’t voice, it is data. Vodafone Europe, for example, has reported 100% growth in data volumes in its 2009/10 financial year, to take data to 77% of traffic volume, despite accounting for only 12% of revenue. For NTT DoCoMo, over 90% of the traffic is data. A recent 4Ggear report on 4G infrastructure trends has suggested that per-capita mobile data usage will grow by an enormous 10,000% in the next five years. Put simply, we need LTE.

It is clear that the existing macro networks cannot support this growth (some would argue that they barely support the current demand), which ultimately means that the fundamental network architecture needs to change for LTE. In any wireless system, the best performance is achieved by close to the cell site: the further the user moves away, the faster the performance drops off. This is where small cells come into play: short range equals good SNR (signal to noise ratio) which equals high data rate. Long range means bigger cells, each of which can serve more people (a good thing) – but the trade-off is that available data rate per person falls (a bad thing). An operator launching a service predicated on ‘high-speed’ needs to ensure that users are consistently within range to receive the advertised data rates and that it provides enough capacity per user overall.

Two of LTE’s attractions are its ‘flat architecture’ and self-organizing network (SON) capabilities. Both of these features make basestations a lot simpler to install and manage: both capital expenditure and operating expenditure therefore fall. In other words, with LTE it is both desirable and feasible to deploy lots of small basestations. (When 3G was first deployed, femtocell technology had not yet been perfected. So, although the same logic in principle applies to HSPA+ small cells or ‘metro femto’, in practice operators face a different equation in adding small cells to their networks, rather than building with them from scratch).

High-volume production of basestations brings economies of scale: and the relentless progress of technology towards reference designs, third-party software and standardized chips (system on chip, or SoC) drives down costs dramatically. So both OpEx and CapEx trends favour the widespread deployment of a dense network of small cells.

Utilising small cells to support the macro network can deliver high data rates to subscribers without requiring hugely expensive investment in new base stations. However this move to small cells represents a highly disruptive change for operators and equipment suppliers alike: it enables new business models, but also requires a dramatic change of thinking.

The major challenge with LTE is migrating from the current infrastructure and how best to support this. This migration is something that operators across the world are doing already, installing LTE base stations in existing 3G locations because it’s a quick, easy, and cheap way to achieve broad coverage. However, operators are starting to realize that simply providing wide coverage isn’t going to be enough to support the required demands. Small cells are instead needed to support the macro network, with targeted femtocell deployments in specific enterprise, rural and residential areas. In fact ABI has forecast that by 2014 shipments for LTE femtocells will reach 20 million annually, with most LTE femtocells supporting 3G in multimode.

There is also an opportunity for WiMAX providers to use femtocells to make the move to LTE. Many WiMAX infrastructure providers have shown strong interest in expanding their product portfolio to address the LTE opportunity as an ‘addition to’ rather than an ‘instead of’ approach. This will allow their operator customers to expand their market potential by adding LTE capabilities. Small cells are therefore a great way for infrastructure vendors to approach operators to break into the LTE market.

The next question for a vendor, therefore, is whether to develop small cell technology in-house, or look to buy it in. Fortunately, telecom infrastructure industry standards have evolved to the point where it is no longer necessary for network equipment providers to design, build, or even integrate systems in-house. In fact, with network element functionality progressing to the point of being innovative application software hosted on a standards-based platform, the strategic rationale for an equipment provider to build its own set of femtocell devices is actually declining quickly.

Instead, infrastructure vendors can focus in-house resources on unique value add – in other words, on application development – and turn to the telecom ecosystem for complete, standards-based femtocell development systems on which the applications are run. A good example of this is the LTE reference design developed jointly by picoChip and Continuous Computing. This will help accelerate time to market for product developers, femtocell access point manufacturers, network equipment vendors and operators. The reference design is the industry’s first to include the LTE modem (PHY), radio frequency (RF) and packet processors, protocol software, intelligent router functionality and a complete Evolved Packet Core (EPC) simulator. It is an excellent way to accelerate and de-risk development so that a large team or upfront investment is not needed; and it allows WiMAX vendors to migrate to LTE with minimal risk and avoid duplicating effort already spent on WiMAX development.

The future of the cellular industry is without doubt one based on LTE. With more tablet devices and smartphones being released every month, wireless networks simply have to evolve and adapt in order to deliver the services that customers want. Small cells will enable this shift, not only providing coverage in congested areas such a stations and city centres, but giving those with existing 3G infrastructure a cost-effective way of upgrading their networks, and opening up the market to WiMAX vendors without risking extensive and heavy investment.