Femtocells, or tiny cellular base stations, are one of the most talked about technologies in the mobile industry today. And with hives of femtocell activity at many of the world's biggest operators, Light Reading has compiled a comprehensive list of the companies targeting the femtocell market and their products.

For the record, a femtocell is a "low-power wireless access point that operates in licensed spectrum to connect standard mobile devices to a mobile operator’s network using residential DSL or cable broadband connections," according to the industry group Femto Forum Ltd.

The pint-size base stations go into homes or offices to improve indoor cellular coverage and the traffic is backhauled to the mobile core network over users' own broadband connections.

It is not yet exactly clear whether many customers will actually pay for the diminutive RF helpers. Most of the femto activity this year has been in operator labs, field trials, and standards bodies. The number of operators with live, commercial femtocell services today can be counted on one hand, namely Sprint Corp. (NYSE: S) in the U.S. and StarHub in Singapore.

This "Who Makes What" report covers the basic CPE and network equipment operators will need to deploy femtocells: the femtocell access points – the actual devices that are installed in homes and offices; femtocell gateways – the network elements that aggregate femto traffic; and the chips that go into the femtocell access points.

Areas such as network management and service provisioning have not been included in this report, even though these systems will be important for operators to determine the feasibility of a femtocell deployment and choose from among competing vendors' solutions. These issues can be addressed at a later time when more operators are actually deploying femtocells commercially and the management systems will take on greater significance.

As with previous "Who Makes What" reports, we ask you to help us complete the picture. If there are companies or products that we missed or if there is information here that is inaccurate, please tell us on the message boards or send an email to [email protected], and we'll update the information. (See Who Makes What: Telco Home Gateways and Who Makes What: GPON Equipment.)

Here's the hyperlinked menu for the femto feast in this report:

— Michelle Donegan, European Editor, Unstrung

Next Page: The Small State of Play

Femtocells promise to improve indoor cellular coverage, increase network capacity, and integrate mobile handsets into home networking environments, all of which will save costs, reduce churn, and increase service revenues from new mobile data services.

That's a lot to expect from a little base station.

The femtocell proposition certainly sounds compelling and operators are now testing the ambitious assumptions of the fledgling femto market. The technology has even attracted the notice of Google (Nasdaq: GOOG), which got involved in 2007 through a strategic investment in U.K. femto startup Ubiquisys Ltd. (See In Focus: Femtocells, Google Invests in 3G Startup, and UbiquiSys Gets Google Boost.)

The only commercial services in operation so far are from Sprint Corp. (NYSE: S) in the U.S. and StarHub in Singapore. Japan's SoftBank Corp. is next in line for a commercial launch since it announced plans for a service to start in the first quarter of 2009. (See Sprint Goes Femto, StarHub Launches 3G Femto Service, NEC, Ubiquisys Win Softbank Femto Deal, and Japanese Femto Rules.)

News of equipment contract awards started to trickle in at the end of 2008. Softbank announced that it will deploy NEC Corp. (Tokyo: 6701)'s solution, which incorporates the femtocell from Ubiquisys. And Alcatel-Lucent (NYSE: ALU) announced that it has won a deal with Chunghwa Telecom Co. Ltd. (NYSE: CHT) in Taiwan. Unstrung also exclusively reported that NEC and Huawei Technologies Co. Ltd. won a tender for enterprise femtocells at Orange (NYSE: FTE)'s Orange. (See Chunghwa Femto Tender, Chunghwa Picks AlcaLu Femto, and Huawei, NEC Score Orange Femtocell Deal.)

In the U.S., Sprint won't be alone for long with its CDMA indoor coverage service as the femto scene heats up in 2009. Verizon has told Unstrung that it could have a femto product on the market in 2009, and AT&T has said that it expects to launch a commercial femto service in the second quarter of 2009. (See Verizon Eyes Femtos for 2009, AT&T Targets Q2 Femto Launch, AT&T the Latest to Jump on the Femto Wagon, Cisco, ip.access Prep Femto Combo, and Femtocells Face Uncertain Future in US.)

But most commercial deployments elsewhere are not scheduled to launch until 2010 or later, according to the Heavy Reading operator survey, "Femtocell Deployment and Market Perception Study." (See 2010: Year of the Femto and Femtocells & Notspots: Coverage Is King.)

Among the 111 responses from the 79 operators surveyed, 54 percent said that they planned to launch services between the second half of 2009 and the end of 2010, and 33 percent said their commercial femtocell launches were scheduled for 2011 or later.

Those timescales indicate that operators still have kinks to iron out in their femto strategies, from the business case to technical issues like interference management, standardization, service provisioning, and OSS/BSS integration. (See Operators Feel Femto Frustration, SFR’s Femto Riposte, and Femtos Turn to DSL for Provisioning Smarts .)

The survey found that operators are most concerned about the lack of standards and femtocell interference with the macro network.

The femto community worked hard to resolve the standardization issue in 2008, and the 3rd Generation Partnership Project (3GPP) finalized the first batch of specifications for the Home Node B (that is, UMTS femtocell) at a meeting in December 2008. Products incorporating the new standards are expected to hit the market by the end of 2009. (See Vendors Unite on Femtocell Architecture and 3GPP Picks Femtocell Standard.)

As for the interference concerns, the Femto Forum has completed a major research project that identified several different interference mitigation techniques that operators can use to minimize femto interference with the macro network when femtocells are deployed in the same radio channel, or carrier. The Forum will contribute this work to standards bodies to ensure standard approaches to interference management. (See Femto Forum Studies Interference and Femto Firms Counter Interference Flak.)

There is also not a clear answer to why operators should deploy femtocells, as business cases remain elusive. For now, the most popular reason to launch a femto service is to improve indoor coverage. But there are other reasons that a business case could be built on, such as increasing network capacity, reducing backhaul costs, and the potential for new revenues from home networking and mobile data services.

Even though technical issues and business case quandaries persist, many operators are forging ahead and putting femtos through their paces. Here are some highlights of operator femtocell projects that Unstrung has covered recently:

Next Page: Femtocell Access Points

Femtocell access points are the hottest little boxes on the cellular scene today.

Most of the devices featured in this report are in operator trials. Among the few femtocells in commercial service today are Samsung Corp. 's CDMA UbiCell, which Sprint brands as the Airave, and Huawei Technologies Co. Ltd. 's uBro Home access point, which StarHub in Singapore recently launched to about 200 customers. NEC Corp. (Tokyo: 6701)'s femto solution, which incorporates Ubiquisys Ltd. 's ZoneGate femto, will debut at SoftBank Corp. in Japan early in 2009.

We've compiled a list of 24 access point products and presented the list in two tables to view the products in different ways: Table 1 details the nuts-and-bolts technical specifications and Table 2 highlights commercial and market position information.

Technical specifications
We've supplied details on some basic technical specifications here for the femto access points: the air interface support, capacity, network interface, and architecture.

The overwhelming majority of products support the 3G technology high-speed downlink packet access (HSDPA) and typically at speeds up to 7.2 Mbit/s. And the nearly universal usage capacity of the devices is up to four simultaneous voice or data sessions.

The network interface refers to the technique used for the connection between the femto access point and the femtocell gateway. Most of the network interfaces in use today for HSDPA femtocells are based on existing standards but proprietary. A new standard – the so-called Iu-h standard - just got the 3GPP stamp of approval in December 2008. Products that support the new standard are expected to be available in 2009. It is hoped that the new standard will foster the nirvana of femto interoperability, where any access point can work with any gateway product.

We've used the term "architecture" to refer to the network functionality that vendors have packed into the access points. There is some variation here among vendors' products. Some vendors have built in basic radio access network (RAN) elements, which include the node B – the base transceiver station (BTS) in UMTS terminology – and the radio network controller (RNC). This architecture is sometimes called the "collapsed RAN" approach.

Other access point architectures include, in addition to the RAN elements, some of the functionality of core network elements, such as the gateway GPRS support node (GGSN), serving GPRS support node (GPRS), and even the mobile switching center (MSC) and home location register (HLR).

For example, Ubiquisys founder and chief technology officer Will Franks, says the company has implemented "every part of the stack" in the ZoneGate, and describes the device as "a complete 3G network in a box."

Other terms used to describe femto architectures are "collapsed stack" or "collapsed architecture," but they are not universally defined and appear to mean different things to different people. But typically, these terms seem to be used to describe an architecture that includes RAN as well as core network element functionality.

{Table 1}

Commercial status
Table 2 provides a snapshot of how femtocell access points are positioned in the market today in terms of commercial availability, trial activities at operators, and partnerships with other vendors.

Where vendors have said their access points are commercially available, we tried to be more specific, but only a couple of vendors divulged details about product volumes. It's probably safe to assume though that products are available in trial volumes at this point. For HSDPA femtocells, for example, most vendors will be tweaking their access points to comply with the newly approved Iu-h standard in 2009.

Standalone or integrated?
In both tables we included standalone and integrated femtocells. A standalone femtocell is just that – a box with a femtocell inside. An integrated femtocell, or gateway, is a home gateway device that has a femtocell module installed. Integrated gateways will typically include an ADSL2+ modem, WiFi router, femtocell, and Ethernet ports. (See Femto Players Gun for Gateways, Cisco, ip.access Prep Femto Combo, Cisco Invests in ip.access, Thomson Sidelines ip.access, Netgear Gateway Goes Femto, and Pirelli Gateways Go Femto .)

2Wire Inc. , Motorola Inc. (NYSE: MOT), Netgear Inc. (Nasdaq: NTGR), Pirelli Broadband Solutions , and Thomson S.A. (NYSE: TMS; Euronext Paris: 18453) have already integrated femtocells into their home gateway products. And Huawei says that it will offer an integrated femtocell in 2009.

Where possible, we have indicated who supplies the femto module to the home gateway vendor.

{Table 2}

Next Page: Femtocell Gateways

Femtocell network gateways sit between the femto access points and the mobile core network and provide (for lack of a better term) a gateway into the core network. At the most basic level, the job of a femtocell gateway is to aggregate and secure femtocell traffic.

The gateways are needed because mobile core networks were not designed to handle tens or hundreds of thousands of small base stations like femtocells. So, traffic from the small home base stations needs to be aggregated, or concentrated, before hitting the core network.

Another key function of the femtocell gateway is security. Since femtocells use consumers' own DSL or cable broadband connection to backhaul traffic to the core network, that traffic at some point travels over the public Internet. This traffic needs to be secured.

Products in this space are evolving as the femtocell market develops. It is not straightforward to provide an apples-and-apples comparison for different vendors' products. With that in mind, we've compiled a list in Table 3 of network products that target the femtocell gateway market and provided a brief description to illustrate what functions are served by each one.

The product and vendor landscape is diverse. For example, some vendors provide parts of the femto gateway functionality, such as Kineto Wireless Inc. 's UMA network controller or Tatara Systems Inc. 's convergence server. (See Kineto, Ubiquisys Team, NEC, Kineto Partner, and Kineto Releases UNC.)

Several of the femtocell access point vendors, such as ip.access Ltd. , have also developed gateways or access controllers to go with their femtocells.

Large equipment vendors like Alcatel-Lucent, Huawei, and Nokia Siemens have adapted existing network products to create femtocell-specific solutions.

{Table 3}

Future developments in this space will come from vendors heaping additional intelligence onto their gateway products to support advanced services and applications. For example, Tatara wants to integrate applications around femtocells into its convergence server product, so that services like enhanced call forwarding could be supported in an enterprise femtocell deployment. (See Tango, AirWalk, Tatara Team.)

Next Page: Femtocell Chips

2008 saw a couple of big developments in femtocell silicon that showed just how confident some chip makers are that the tiny base stations will take off.

The first was in June, when femtocell specialist picoChip raised the bar for low-cost optimized femto chips with the introduction of a new family of system-on-chips (SoCs), called PC3xx, that is designed to reduce the total bill of materials by between 50 and 66 percent. The first product in the series is for HSPA femtocells. (See PicoChip Unveils Low-Cost Femto Chip.)

Femtocells are plagued by a classic chicken-and-egg scenario: Operators want the cost of equipment to come down before placing volume orders, but chipmakers need to see the volume orders before committing R&D money to developing the lower-cost silicon. PicoChip's announcement in the summer was a first stab at breaking that deadlock. (See Femto Chips Too Costly.)

The magic price that operators seek is under $100 for a femtocell. But the industry is some way off from that. According to the Unstrung Insider report, "Femtocells: U.S. Market Prospects," femtos aren't likely to cost less than $100 until after 2010. The report estimates that the current total femtocell cost per port is about $160, and the bill of materials is $105. (See Vodafone CEO Seeks Cheap Femtos.)

PicoChip is understood to be working on the next products in the PC3xx family, but the company has not disclosed details yet. Future products are believed to include further development of WCDMA (such as more users supported) and other air interfaces, as well as dual-mode LTE and HSPA.

TI wades in
The other big news was in December when Texas Instruments Inc. (NYSE: TXN) officially got off the femto fence and revealed its own femtocell product. TI is the first major chip vendor to back the little base stations. To have the chip giant throw its weight behind femtocells was hailed as fantastic news for the fledgling market, but femto insiders were bemused. (See TI Dives Into Femtos.)

TI's commitment to femtocells appeared half-hearted to some. The chipmaker said that while it was committed to femtocells, it would not develop optimized silicon now because the femto market was too fragmented. The company said such specialized silicon will be needed in 2010 or later and not before.

We've listed femto silicon companies and their products in Table 4. Since most are startups or privately held firms, we provided some information about the company's investors too.

{Table 4}

There are several companies worth mentioning here that aren't included on our list. For example, Qualcomm Inc. (Nasdaq: QCOM) is understood to be working on a femto chip and has made a strategic investment in ip.access Ltd. . And Israeli mobile WiMax silicon pioneer Runcom Technologies Ltd. plans to develop WiMax and Long-Term Evolution (LTE) femto chips. (See Qualcomm Invests in ip.access, LTE Hopeful Runcom Pockets $10M , and T-Mobile Backs Femto Chip Startup.)

Femto vendors have various strategies for sourcing silicon. For example, Airvana Inc. uses off-the-shelf commercial components and adds its own software. Huawei Technologies Co. Ltd. and Samsung Corp. make their own chips. And RadioFrame Networks Inc. designs its own silicon for its femto devices, but plans to sell its second-generation OmniRadio processor to third parties around 2010. (See Femto Vendor Lands $28M.)

Next Page: The Femto Future

The hottest future developments for femtocells will be in enterprise devices, 4G mobile broadband technologies LTE and WiMax, and so-called metro femtos.

The femtocell proposition is not restricted to consumers at home. Orange (NYSE: FTE)'s Orange revealed to Unstrung that it is planning a femtocell deployment for small-to-medium size businesses and has selected enterprise femto access points from Huawei Technologies Co. Ltd. and Ubiquisys Ltd. (which is part of NEC Corp. (Tokyo: 6701)'s solution). (See Huawei, NEC Score Orange Femtocell Deal.)

For Orange, the business case is about improving indoor 3G coverage. But there is also an argument for increasing data network capacity through enterprise femtocells. The femtos can also be integrated with certain IP PBX functions, so mobile phones could have access to corporate directories like a desk phone.

Alcatel-Lucent (NYSE: ALU) is also working on an enterprise femtocell.

In terms of hardware, the femto access points for consumers and enterprises are essentially the same. But for the enterprise, the devices are beefed up to handle more users and are enabled so that there can be call handoffs between them. So, a small business with poor cellular coverage could have a network, or cluster, of several femtocells in its building or on its campus.

4G femtos
Comcast Corp. (Nasdaq: CMCSA, CMCSK) stirred up a storm about WiMax femtocells when the cable operator revealed earlier this year that it intended to deploy a WiMax femto network as part of its deal with Clearwire. Following this revelation, WiMax femtos were suddenly all the rage. Comcast appears to be the trailblazer for this flavor of femto, but now that the Clearwire asset merger is complete perhaps the cable operator may not be alone for long in its pursuit of better indoor coverage. (See Comcast Goes for WiMax Femtocells and Sources: Comcast Wants WiMax Femtos in 2009.)

For LTE, it is believed that femtocells will be an integral part of early network rollouts. One theory has it that LTE deployments could be femto-led – as in, the femtocells are deployed first in people's homes or public hotspots and the 3G network is used for macro coverage. Then the LTE macro network is built up as more capacity or greater coverage is needed. (See PicoChip Unveils LTE Femto Design.)

Metro femto
Some operators want to take the femto notion further afield for LTE. They would like to take femtocells out of homes or offices and onto the streets in a metro femtocell network. The concept of metro femtocells is gaining momentum and supporters after Vodafone Group plc (NYSE: VOD) shared its vision of an outdoor, self-orangized femtocell network in July. (See Vodafone Dreams of Metro Femto, Operators Eye LTE Metro Femtos, and picoChip Touts LTE Femto.)

Femto chip maker picoChip has a more radical approach. The company believes that all LTE base stations deployed will be small femtocells. The nature of LTE technology calls for more cell sites to achieve the full capacity improvements, and that means the base stations will have to be small. Think femtocells on lamp posts.

A host of other issues such as service provisioning, network management, implementation of the TR-069 standard, and developing femto applications and services will also be keeping the femto industry busy this year.

The femto frenzy isn't likely to calm down in 2009. But as more operators complete their equipment tests and customer trials in 2009, the industry should have a clearer picture of how big the femtocell market is really going to be.

— Michelle Donegan, European Editor, Unstrung