Beyond fixed/mobile convergence

Designed to support FMC services, femtocells are set to transform the way wireless networks are built

IT'S BEEN A FEW YEARS since a telecom technology has come along that deserves to be called “disruptive.” But if we use the term to describe something that provides more than incremental performance improvements, instead changing the rules of the game, then femtocells may well deserve the moniker.

Femtocells are scaled-down cellular base stations designed for installation at the customer premises, where they connect to DSL, cable modem or other broadband connection. The first U.S. femtocell deployment has come from Sprint, which uses the technology to support a fixed/mobile convergence (FMC) offering dubbed Airave. Outside the home, the customer's cell phone operates on the conventional cellular network and is billed accordingly. Inside the home, the voice call or data connection is handed off to the femtocell, which relies on the DSL or cable modem to provide backhaul connectivity. Because airtime charges are avoided, customers can make unlimited calls while they are at home and operating on the femtocell. Customers can entirely forego the cost of a landline phone and enjoy the convenience of having a single phone number for calls in and outside the home.

Femtocell developers are seeing strong interest on the part of wireless carriers, large and small. “We see [requests for proposal] continually coming our way,” said Serge Pequeux, president and CEO of femtocell manufacturer AirWalk. “Most operators are trying to build a model around the femtocell and see its value.”

From the network operator's point of view, femtocells already pack considerable power by generating substantial savings on backhaul costs and by enabling cell phones to work in homes without cellular coverage, which, by some estimates, includes one in 10 cellular customers. Moving forward, femtocells are poised to enable even more powerful applications and to play a vastly expanded role as network operators build out fourth-generation wireless networks.

THE COVERAGE AREA of a femtocell is in the tens of meters, and its cost is in the $100 to $200 range. Sprint, which declined to be interviewed for this article, apparently is subsidizing part of the cost of femtocells in its deployment, charging customers just $50 for the device plus a monthly charge of $15 for an individual or $30 for a family. The carrier likely recovers the subsidy quickly through savings in backhaul costs, in lower churn and potentially in lower acquisition costs.

Depending on the broadband pipe to which it is connected and the type of device used, a typical femtocell can provide bandwidth in the range of 10 Mb/s, which can be shared by as many as eight users. Sprint has not released take rates for its offering, which supports up to three users, but operators expect the femtocell's multiuser support to generate an increase in family plans and to help convert some family members from other wireless operators.

Because Sprint's femtocell offering is based on second-generation 1XRTT service, customers do not get maximum data rates today — and the company's promotional materials barely touch on data, instead emphasizing voice capability. But as wireless operators begin to use the technology with later-generation wireless networks, data rates supported will increase, and data capability is expected to become increasingly important.

“To create a strong business case, data has to be in the mix,” said Roberta Wiggins, a research fellow in service provider mobile enabling technologies for Yankee Group. As consumers become more and more accustomed to using the data capability of their cell phones, they will expect that capability to work with femtocell service, Wiggins said.

Data also will become increasingly important as carriers begin to deploy fourth-generation wireless networks based on long-term evolution (LTE) and mobile WiMAX — and some industry observers believe femtocells will play a more prominent role in those networks than they do in current ones.

These observers note that usage of cellular devices inside homes and offices already is on the rise, and that the trend will accelerate as higher data rates become available. But getting adequate indoor service can be more challenging at higher data rates.

“The newer standards tend to be at higher frequencies, where the signal doesn't go through walls so well,” said Rupert Baines, vice president of marketing for femtocell developer PicoChip. Although penetration is pretty good at 700 MHz, he said higher frequencies are problematic when large amounts of data are involved.

Operators' second- and third-generation network deployments started with macrocells and huge towers capable of covering 20-mile areas, then more infrastructure was added later to fill in heavily used hot spots. But fourth-generation networks may be constructed quite differently. “LTE and WiMAX will launch more efficiently and sooner with smaller cells,” Baines said.

Simon Saunders, independent chairman of the Femto Forum, an industry group aimed at promoting the growth of femtocell technology, agreed that ensuring adequate indoor coverage could be a problem in fourth-generation networks. “To deliver consistent coverage inside soaks up capacity,” he said. “To cover a user inside, you have to sacrifice capacity that you would have been able to offer to 10 users outside the home.”

If instead a network operator were to deploy a femtocell inside the home of an early 4G adopter, “suddenly 10 outdoor users now have better service,” Saunders said.

“Instead of rolling out tens of thousands of cell towers, operators can build their 4G networks from the inside out,” Saunders added. “Users who want service early will need it in their home and can fall back to 3G outside the home. You can create islands of the best service you can offer and good 3G elsewhere.”

The economics of femtocells, which shift power, real estate and backhaul costs to end users, are so compelling that operators may opt to foot the bill for them, eliminating the upfront femtocell charge for consumers, said Will Franks, chief technology officer and co-founder of femtocell developer Ubiquisys.

“We've had carriers say they believe femtocells are fundamental in LTE networks from Day 1,” Franks said. “LTE may lead with femtocells because it enables operators to target LTE coverage where it's really needed.”

MOVING FORWARD, femtocells also have the potential to enable some appealing value-added services based on a wireless network's ability to know when a call is handed off from the nearest macrocell to the femtocell. Parents could receive an alert when a child arrives home from school, or photos that a user took with his or her cell phone could automatically be downloaded to a home computer when the user steps in the door.

The tighter integration of the cell phone with other home systems also could give rise to creative new services. A set-top box with a built-in femtocell, for example, could lead to an application that would automatically ask end users watching a football game if they'd like to order a pizza prior to half-time, Baines said. He added that integrated devices could reduce the cost of premises equipment by as much as 30%.

Because femtocells eliminate airtime concerns in the home, they also may drive even heavier data usage. Saunders noted that an end user might watch video programming on a wireless device if there were a conflict between what various family members wanted to watch.

The low cost of mobile data service within the home could drive end users to try services they might not otherwise use, Franks said. “Consumers are often scared of data because they don't know what it will cost,” he said. But in their “home comfort zone,” he added, “they will be willing to try things out.”

Although the earliest applications of femtocell technology have been residential, some manufacturers see significant potential in the business market and have developed or are developing products to suit that space.

“The major operators are looking at the enterprise business as an opportunity to expand the network,” Pequeux said. He noted that AirWalk has developed a femtocell targeting the enterprise market that costs between $3000 and $4000. That product supports up to 32 users and has the added benefit of allowing PBX features to operate within a campus area. The device also supports handoffs between femtocells, enabling capacity to be expanded by deploying multiple units within an enterprise.

FEMTOCELLS ARE NOT the only choice for network operators wanting to offer FMC. The chief rival technology in this area uses dual-mode Wi-Fi/cellular handsets to deliver a similar service.

Such offerings eliminate the need to purchase and install a femtocell, relying instead on Wi-Fi routers already installed in many homes and businesses. As with femtocells, in-building calls are shifted onto a landline broadband connection, in this case via the Wi-Fi router.

The downside to dual-mode offerings is that because Wi-Fi operates on unlicensed frequencies, it may be more prone to interference. Another concern is that customers require a new handset with a rather hefty price tag of around $200 to $300. Because fewer manufacturers offer these devices, feature sets also are somewhat limited. Some see the handset as a major obstacle to dual-mode deployments. “As [Robert] Metcalfe said, ‘You should always bet on the installed base,’” Baines said.

Dual-mode offerings may prove to be most popular with carriers that don't have their own wireless spectrum. Embarq, one of the U.S. carriers that have launched a dual-mode service, falls into that category, and another, Cincinnati Bell, has spectrum only in its local area. A third carrier offering a dual-mode service, T-Mobile, has extensive wireless spectrum but also has a substantial Wi-Fi hotspot business.

Service providers that have their own wireless resources or have close alliances with wireless carriers are expected to favor the femtocell approach because it gives them more control. When dual-mode customers roam on Wi-Fi, the carrier loses valuable knowledge about customer connectivity — and because customers are no longer using the cellular network, they may not be able to get some of the advanced service features they would get from a femtocell.

Femtocell technology hasn't been perfected yet. Some industry observers have expressed concerns that in certain situations femtocells could interfere with neighboring macrocells. But developers are working to address those concerns and are expected to devise a solution.

“Using auto-configuration, you can adjust the power level so it won't interfere,” Wiggins said.

Another concern relates to the portability of a device that is intended to be part of a service provider network. “If someone takes it out of its home area, you have to make sure they won't have access to spectrum,” said Peter Jarich, research director for Current Analysis.

Some potential problems are more political than technology-related. When the operator providing the landline connection is not under the same ownership as the femtocell operator, there undoubtedly will be room for friction — particularly if the broadband carrier perceives a loss of revenue or increase in costs as a result. But Vonage and other voice-over-IP operators that have faced similar issues have won most of the regulatory challenges against them — and most industry observers expect to see similar results for carriers deploying femtocells.

By using spectrum more efficiently, femtocells will increase the spectrum's value, which tends to make regulators happy, Saunders said. Besides, he added, “Lots of users getting great service at a low cost won't be a negative thing.”