With the end of the year fast approaching, telecom analyst "prediction pieces" are easy to come by. The same holds true as bloggers, journalists and anyone else with an opinion on the market (or an agenda to push) dusts off their crystal ball and takes a look inside. If you don't want to be too daring, a crystal ball isn't even necessary; we can sketch out the technology trends in the wireless infrastructure space with little trouble. Based on conversations with operators and vendors, we know what demos to expect next February at Mobile World Congress. We know the technologies that will get most deployed--and those that will still face an uphill battle. The general trends are already set.
Most of them, however, I don't plan to talk about here. For now, I want to focus on a rather narrow question: how will LTE impact the femtocell space?
The question is particularly interesting for a few reasons. First, there is the once-popular supposition that LTE - as a technology break from the past - would be built from the "inside out." This hasn't materialized, but the rationale is obvious; if LTE is about delivering solid data capacity, the use of femtocells is one way to obtain solid spectrum reuse and bandwidth per user. Second, there are the product demos and announcements. picoChip, most notably, has been crowing about its LTE-ready femtocell silicon for most of this year. Vendors like NEC and Kyocera are building LTE strategies around small cells, with NEC telegraphing plans for its own LTE femtocell R&D. Whether or not LTE femtos or other small cell get deployed next year, there can be no doubt that we'll hear a lot about these products. Finally, there is a recent survey we ran.
Last March, we ran a survey intended to provide an indicative (rather than statistically significant) set of insights into the femtocell ecosystem based on a small sample of key industry players. Our latest survey--conducted in October and the early part of November--carried this forward. Key decision makers (marketing VPs and directors, CEOs, etc.) in the femtocell vendor ecosystem were fielded a web survey. With a fairly small sample size (~20 respondents), it would be difficult to qualify our latest study as statistically significant. On the other hand, where the femtocell vendor ecosystem still includes a relatively small number of firms who are actively involved in and shaping the market, our sample captured a majority of the players who matter. Regardless, it provides another view into the market, a view from people who are, arguably, closest to the market's development.
What the vendors are thinking
LTE was not the sole topic of our survey. In many ways, however, it was one of the most interesting aspects, if only because it is one of the areas where shifting responses suggest a change in the market's views.
We might question how quickly operators will move on small cell LTE. We might question how broadly they will deploy small cells like picocells or microcells, but operator messaging on LTE small cells has generally been positive, and this is reflected in our ecosystem survey; the vast majority of femtocell vendors see small cells as "critical" to operator LTE launches. Less certain, however, is the role of femtocells in LTE. Here, the opinions were essentially split down the middle--with a slight edge (bias) going to the idea of femtocells playing a major role in LTE deployments. More interestingly, however, is the shift in perceptions from March when 71 percent of respondents saw a clear, critical role for femtocells in LTE. While not a massive shift (15 percentage points) it does align well with the fact that initial LTE deployments are already underway and LTE femtocells are not yet commercial.
If commercial availability is a gating factor, however, the obvious question is when can we expect commercially available LTE femtocells...then when will they get deployed?
On the topic of commercial LTE femtocell availability, one thing is clear: today's femtocell vendors expect a year delay between product availability and service rollouts. A majority of the vendors surveyed expect commercially available LTE femtocells in 2012; a vast majority (77 percent) expects them in the 2012-2013 timeframe. Commercial deployments, then, are expected by a majority in 2013, with a vast majority planning for 2013-2014. More than once, Japan was singled out as an important driver--in line with aggressive moves on LTE in the country and traditional issues with in-building coverage.
What does it all mean?
The question of 4G femtocells (yes, I just called LTE 4G), is an enduring one. Operator interest in controlling user experience and maximizing the use of their spectrum argues in favor of them. The ubiquity of WiFi in smartphones combined with the cost efficiency of 802.11 argues against them. With today's 3G femtocell services being largely marketed as a voice coverage solution, the low priority that operators have put on Voice over LTE (other than circuit-switched fallback) in the near-term suggests that LTE femtocells won't be an operator priority for another two or three years. And when they do come, operators will need to find a way to make them compelling, doing more than just controlling the connection but also controlling the experience. Perhaps they'll finally get around to rolling out those femto-zone applications we've heard about for so long. The relatively even split in opinions on the importance of femtocells in LTE reflects these arguments and uncertainty.
At the same time, the massive discrepancy between expectations between LTE femtocells and non-residential LTE small cells makes a few points clear. Few vendors in the femtocell space are strictly femtocell vendors--they see opportunities to push beyond the home or enterprise. They may fight vendors with more macro network expertise, but they see an opportunity (today with 3G, tomorrow with LTE) to be tapped. Too often the concept of the "femtocell" gets conflated with a variety of others--picocell, microcell, "small cell" macrocells--though most vendors believe the key differences relate to cost, capacity and coverage per our survey.
Though these are obvious differences, it may be surprising that issues such as self-organization, open access and handover weren't more cited. Publicly, most femtocell vendors--at least those without traditional macrocell businesses--have argued that they see no conflict between their deployment and interference mitigation tools vs. the SON tools being rolled out; suggesting that these should be distinct solutions. At the same time, macrocell platforms--and many picocells--will need to be open access, servicing any user on the operator's network. Femtocells shouldn't fall into the same usage scenario. Finally, while it's assumed that operator deployed solutions like picocells and macrocells will need to support both hand-in and hand-out, this won't always be the case with femtocells.
While LTE femtocells are coming, the safer bet is on the more generic small cell used to fill in coverage and capacity gaps or hotspots; it's nearly certain that operators will deploy these solutions. To tap the opportunity, however, vendors need to think long and hard about the differences between these platforms and how best to market their expertise outside the home.
Peter Jarich is the Service Director leading Current Analysis telecom infrastructure practice. Follow him on Twitter: @pnjarich.