Femto Fray: The Obstacles Ahead
Type: Advisory Report
Analyst: P. Jarich
Report Date: October 30, 2006
Module: Wireless Infrastructure
ID: CIR23633 |
|
Summary
Issue
The basic femtocell concept is relatively easy to grasp: inexpensive, residential base stations designed to deliver most of the functionality of current Fixed Mobile Convergence (FMC) solutions – superior indoor coverage, spectrum offload from the wide-area mobile network, transport offload from the wide-area mobile network, improved data rates in the home, service stickiness and the potential for innovative tariffs – all without relying on a limited selection of (often) expensive devices with WiFi built-in.
Yet, if femtocells offer all the benefits of FMC, plus the ability to leverage off-the-shelf 2G or 3G devices, why haven’t they been deployed en masse yet? In the past, cost has been the major sticking point. Squeezing the bill of materials on a wireless base station down to consumer pricing ($100 to $200) is no small task. Vendors, however, claim that this problem has been solved – or, at least, can be solved thanks to silicon innovation and volume manufacturing.
If this is correct, the question remains: what obstacles stand in the way of major, wide scale femtocell launches? More than simply pointing to the issues that vendors will need to address as they attempt to take femtocells from vision to reality, a (relatively) exhaustive list is critical for any vendor looking seriously at the femtocell space, preparing them for the work that lies ahead.
Current Perspective
Like any new industry, the femtocell space faces a number of obstacles. If it didn’t, we’d all be enjoying crystal-clear voice calls and 3G data rates in the comfort of our homes – while fretting every now and then over what we’d be losing if we switched to a new mobile provider.
While someone who is dubious of the femtocell value proposition could devise an endless list of weaknesses, 12 key issues stand out as the major considerations facing would-be femtocell operators and vendors. These considerations, in turn, can be grouped into three broad categories: technical, business/strategic, and marketing/regulatory.
• Technical: The femtocell itself is a new approach for a cell site – a product that can be purchased by consumers and installed in the home to meet their wireless demands going forward, while integrating with an operator’s mobile core over the consumer’s broadband service. Development of a consumer electronics device aimed at meeting these demands is clearly tricky. Vendors claim to have done just that, but issues still remain in bringing the product to market.
• Costs: Vendors may claim to have femtocell products which hit the $150 price point. It may be a challenge, however, for vendors to obtain the manufacturing scale to support this price in the near-term – at least without a major upfront operator commitment. Operator commitments, in turn, may be difficult to secure before vendors can prove that inexpensive products are viable.
• Upgrades: Femtocells must be upgradeable in order to support market demands as well as mesh with the operator’s plans for their macro network: HSUPA, EDGE evolved, etc. Macro base stations have been designed to be relatively easily upgraded (SW or module level). Femtocells will need to rely on software-defined radio or DSP solutions. However, this flexibility could impact costs over dedicated ASIC-based implementations. The need to support multiple technologies within one box simultaneously (GSM/UMTS) would be even more burdensome…all conflicting with a need for inexpensive solutions.
• Deployment Tools or Zero Configuration: It is in an operator’s interest to ensure that their femtocells are deployed correctly – providing adequate coverage, networked with the customer’s broadband service (e.g., cable and DSL), seamlessly attached to the operator’s core. If they are not deployed effortlessly and correctly, service reputation (and adoption) will suffer. Clever implementations may be able to dynamically listen and adapt to changes in the radio environment while automatically finding and linking with the operator’s “mother ship.” Perhaps, software-based deployment wizards will be required on this front. The WiFi industry provides a precedent. However, as a product operating in licensed spectrum and connected to the operator (from a network and psychological perspective), femtocell deployment tools will need to be more robust both from the consumer perspective and the operator’s customer service vantage point.
• Technical Details: Beyond driving down price, vendors must tackle a number of technical questions when building their femtocell products. What form will the product take (standalone, integrated into cable/DSL modems, etc.)? How will clock synchronization take place for CDMA products (GPS, IEEE 1588, macro network input)? Where will authentication and authorization take place (in the network, in the device with a SIM)? Clearly, there are answers to all of these questions. Each answer, however, comes with its own costs and tradeoffs.
• Business and Strategic: Assuming that vendors can actually deliver a femtocell that meets pricing, usability, and functionality requirements, operators still face a number of questions when building out a femtocell network and femtocell service. After all, a successful service launch requires the operator to meld a dense network of femtocells with their mobile core and radio access networks while figuring out the right way to actually get the residential base stations into the hands of hungry customers and handle the inevitable set-up and support problems.
• Network Integration – Core: Ultimately, an operator’s femtocell network must be connected into its mobile core, if only to deliver a common set of services and interconnect the femtocell user to the PSTN and people on the operator’s macro network. Several options are available: UMA, IMS, IP Iub into the RNC, or an existing RNC combined with some sort of gateway. Each option, of course, comes with tradeoffs in terms of scale (proprietary gateways and UMA) and/or time to market (IP Iub and IMS). The best option will depend on an operator’s network assets and network evolution roadmap.
• Network Integration – RAN: Maintaining orderly and clean spectrum is essential for sound operation of CDMA and WCDMA networks. Operators must be assured that femtocells don’t destroy outside macro network noise environments and that they can co-exist with other femtocells – which might be installed at a neighboring home or apartment. Use of distinct CDMA carriers will help to mitigate interference with the macro network. Sophisticated use of different pseudonoise codes, power control, and antenna technologies will be needed to deal with networks which are closer in.
• Capacity Planning: Using separate carriers in the femtocell and macro cell network may help to mitigate interference between the two. However, operators who are tight on spectrum (or plan to add macro cell carriers at any point) will want to reuse the same carrier across both – implying the need for deeper coordination, power control, and antenna technologies.
• Consumer Distribution: Ultimately, any operator considering a femtocell launch needs to consider one simple question: where will an end-user get his of her femto? The operator? The vendor? An electronics shop or mail order? Every operator will have their own preferences – preferences which may include a mix of channels. Regardless, each channel strategy assumes a tradeoff. Selling direct, for example, allows an operator to better control distribution and take control of the user experience. Working with consumer electronics outlets, on the other hand, may allow for broader distribution. Pursuing both channels may spread resources too thin in a still nascent market.
• Customer Support: If a product has the operator’s name on it, customers will expect support when it breaks down. Luckily, WiFi has set an expectation of limited support for gear that resides in your house; operators can set customer expectations low. Yet, if the femtocell cannot deliver robust services (strong coverage and reliable service) the value to customers will be minimal. And if they don’t leverage their femtocells, the value to operators will be lost. At a minimum, however, network monitoring tools will need to see down into the femto in order to diagnose problems.
• Marketing and Regulatory: If a technology or service is good enough, it should sell itself. Unfortunately, few technologies are that good and marketing efforts are required to make them into a commercial success. And, even if the theory behind femtocells is sound, operators will want to know that it’s more than just theory while battling the health biases that have dogged 2G and 3G for years.
• Radiation: With 2G and 3G services, users often protest when a new tower goes up in their neighborhood. It’s not unreasonable to expect some level of trepidation when people are given the option to install a base station in their home. In reality, a strong femtocell signal is possible with a very low power level and should help ratchet down handset power, keeping radiation levels ultra low for users. The result is a safer wireless environment. Operators will still need to go on an education campaign if they intend to convince users.
• Net Neutrality: In the U.S., network neutrality continues to be a thorny issue: operators who provision a broadband connection feel a right to give priority to their applications. Vonage and Skype provide popular examples. Where broadband providers with 3G partners or their own spectrum will be at an advantage, others looking to route femtocell traffic over DSL or cable networks may be in for a battle.
• Proof Points: With all due respect to vendor claims, operators will want to see real world examples of working femtocell solutions before jumping on the market. Short of finding a pioneering operator who is willing to take a chance on unproven technology, vendors will need to secure trials that they can point to.
Recommended Actions
Vendor Actions
• Femtocell vendors must be prepared to support flexibility where multiple deployment options exist (e.g., core network integration, form factors, deployment and management tools, etc.). In each case, there is no “right” answer for how to deploy a femtocell service. A vendor may believe that one option is better than the other – but they need to come to market with multiple tools in order to deliver what operators want and drive femtocells into the market.
• Vendors need to be upfront about the bill of materials behind their femtocells. Obviously, no vendor will want to provide complete insight into their products – doing so sheds too much information on sensitive issues like margins. However, to convince operators that femtocells can be delivered at attractive terms through the long-term, vendors will need to share some information on the costs (today and in the future) of their products.
• Femtocell vendors must get working examples of their gear up and running in the near-term. Again, the basic value proposition of femtocells is relatively easy to grasp. However, operators have seen their share of technologies that held a lot of promise but never materialized thanks to business and technical difficulties. To demonstrate that femtocells work and help operators understand the business models driving them, trials and proof-of-concept networks need to be deployed as soon as possible.
• Any femtocell trials or proof-of-concept networks should highlight the ability of a vendor’s products to support technology evolutions going forward. Femtocell upgrades will be necessary, whether to support new core network integrations (proprietary today, UMA or IMS tomorrow) as well as new air interfaces (HSUPA, higher speed HSDPA). Proving that these evolutions can be credibly delivered will be key to winning over operators and getting femtocells into the market.
• Best practices insight – technical, marketing, etc. – must be available from all femtocell vendors. To be sure, many vendors will have specific insight into what they think is right when it comes to network integrations, capacity planning, form factors, tariffing, etc. Given the newness of the femtocell market, however, many others will be hungry for whatever insight can help them to deliver successful services – ideally insight driven by initial deployments and trials.
• Would-be femtocell vendors need to proactively explain away the worries related to radiation. Given the uproar often heard about new 2G and 3G tower builds, there should be no question that end-users will be turned off by the idea of putting a base station in their home. This fear, in turn, could stunt market growth. Since femtocells will help to ratchet down the output power on mobile devices, they should yield a safer environment. Vendors must work to make sure that operators and users understand this point.
User Actions
• Operators must be prepared to offer vendors a major volume purchase in order to obtain the pricing necessary to drive a successful business model. It’s obviously risky to commit to purchasing a large quantity of gear from start-up vendors in a nascent market. However, without this type of commitment, no vendor can hope to hit the femtocell quantities which lead to a $100 to $200 price point.
• Operators looking to deploy a femtocell solution should consider waiting for IMS-based integration solutions to emerge. Proprietary and UMA-based solutions may be available today. UMA solutions, however, will require specific functionality on the femtocell while IMS solutions – likely some form of packet data gateway – will mesh better with more operator core network plans going forward. If there is a need to jump early, operators should ensure that proprietary solutions can evolve to support IMS going forward.
• Operators with 2G and 3G networks need to consider supporting both technologies in their femtocells. Yes, supporting multiple, distinct technologies implies added costs. And, yes, supporting 3G alone may push users to adopt new devices and new services. However, launching femtocells with 2G and 3G in them will support subscriber transitions to 3G without forcing expensive device upgrades. More importantly, until 3G coverage – or dual-mode 2G/3G devices – is ubiquitous, 2G femtos will be necessary in order to deliver seamless macro network handoffs.
• Operators looking at femtocell solutions should consider launching with a hands-off customer support model. If the femtocell itself is branded by the operator, customers will expect the operator to provide extensive support. If, however, operators distribute their femtocells through indirect channels, service support on par with WiFi access points may be acceptable – saving the operator when it comes to customer service staffing and time. |
