consulting, business planning and market analysis on wireless data technologies
 

Subscribe to Senza Fili Consulting Newsletter

Senza Fili Newsletter

September 2004 issue

Welcome to the first issue of the Senza Fili Newsletter.

Alarm receiving station by
Harrington-Seaberg (Moline, Il.)

Much has happened over the last year and a newsletter seems a fitting way to celebrate and to let you know what we are up to and to share our perspective.

Most of our work has been on BWA—Wi-Fi, WiMAX and other non line of sight technologies.

On the mobility and portability front, we have looked at how BWA can support mobile access through roaming and considered which business models can best support mobile wireless access.

On the fixed front, we have focused on how BWA can provide last mile access both in underserved and metropolitan areas and on how service providers, municipalities and communities can benefit from it. We have compared different technologies, to see which are most appropriate for different business models and in different countries.

The more speculative part of the work has been on how BWA will be able to offer both fixed and mobile access over the same network infrastructure and give service providers the opportunity to bundle fixed and mobile access.

The first issue of the newsletter is therefore devoted to BWA. It includes the article I wrote for Total Telecom that gives an overview of the opportunities and challenges that WiMAX offers to service providers. It is preceded by a WiMAX development timeline that reflects the current expectations on technology rollout, services and costs and by a summary of an article on wireless data technologies for municipalities and public agencies.

I hope you enjoy the newsletter—and as always I love to hear your thoughts!

Monica Paolini

The wireless telegraph station at Glace Bay

At a recent meeting on wireless Internet and municipal public safety organized by W2i, I gave a presentation on how BWA can offer a solution to the data connectivity needs of municipalities and public agencies. The presentation has been in part converted to a chapter for a report published by W2i.

The article discusses the advantages offered by BWA technologies and the need for municipalities and public agencies to carefully evaluate the alternatives, in terms of functionality, initial capex, and ongoing operating costs.

In particular, it focuses on Wi-Fi, 802.11 mesh technologies, and WiMAX, which appear to be the key contenders in this market.

September 2004. Adapted from Total Telecom

Heavy hitters like Intel are putting their muscle behind WiMAX R&D, and now the industry is daring to believe that BWA will really prove to be a cost-effective competitor to other broadband technologies.

Broadband wireless access (BWA) is back - but don't look up. New BWA technologies promise small CPE form factors - built into PCMCIA cards, laptops or handhelds - and non-line of sight capabilities, which mean no outdoor antennas and no truck rolls for installations. In addition, they support mobile data and voice connectivity hand-offs up to vehicular speeds and bandwidth throughput that can easily reach 1 megabit per second.

There are still pressing questions about the viability of BWA, centered around return on investment and business models for operators. And there is still much uncertainty about the business case for BWA beyond under-served areas that lack fixed broadband infrastructure.

WiMAX - a point-to-multipoint technology - is widely expected to dominate the BWA market. Client-device support for mobility is still two years away, but WiMAX has already gathered wide industry support. In addition to BWA vendors like Airspan, Alvarion, Aperto Networks and Redline Communications, big hitters including AT&T, BT, Cisco, Fujitsu, Intel, Motorola, Nokia, Nortel and Siemens are all members of the WiMAX Forum, an industry organization that will soon start to certify interoperable WiMAX products.

In the business segment, higher margins for service providers make BWA an appealing opportunity for new entrants in competitive markets. TowerStream, for instance, offers 1.5 Mbps guaranteed connections with additional 3.5 Mbps best-effort bandwidth at $500 per month in Boston, New York and Chicago. Jeff Thompson, CTO at TowerStream, says the company has about 700 customers and has been cash flow positive since June this year.

Business users find that competitive prices - a comparable T1 service costs $600-1,000 per month - and fast setup are major advantages, says TowerStream. For the same bandwidth (1.5 Mbps) using DSL, the cost is $100-$250 per month depending on the provider, but of course users get only a contended service.

Meanwhile, established service providers can widen their coverage area using BWA and avoid more expensive optic fiber installation.

Residential challenges

The residential market, where traditionally margins have been very tight, is a more difficult one. BWA is a scalable technology that enables service providers to increase capacity as and when needed, and this effectively brings down the cost per subscriber. That scalability enables service providers to get initial coverage in an area with limited capacity and at a lower cost. Capacity can be later increased quickly to meet demand.

Fixed technologies like cable, DSL and, in particular, fiber to the home do not have that advantage, because the upgrade or the installation is typically done for entire neighborhoods.

According to vendors' estimates, WiMAX base stations will cost anything between $5,000 and $50,000, depending on capacity and range. CPE prices will also be quite steep to start with - $250-$300, and more for a CPE with an externally mounted antenna - but economies of scale could bring CPE prices down to the $50-$70 price point by 2009.

Wide adoption of WiMAX depends on low CPE prices. Intel is committed to push prices down in the hope that this will replicate the Centrino effect and promote laptop sales. Several vendors already say they do not see the CPE market as profitable and are happy to have Intel mass-produce CPE and let them concentrate on more lucrative base stations.

However, major service providers may decide to roll out services initially only in areas where demand is urgent. They could wait to deploy in other areas once CPE prices have come down. If Intel sees that only rural providers are interested in WiMAX, it may not have too much of an incentive to push prices down much.

WiMAX deployments can be markedly cheaper than DSL or cable upgrades in residential areas. In the US, at a 10% take rate - the proportion of homes passed taking the service - DSL and cable upgrades cost about $1,500 per subscriber, while the projected cost for WiMAX is $700 per subscriber. At a 20% take rate, the figures are $800 and $600, respectively. With a 45% take, costs for DSL/cable and WiMAX are approximately the same.

After that it becomes more cost-effective to install DSL/cable, although it is unlikely that service providers will have that kind of penetration in a competitive market. WiMAX has a clear edge in areas where a service provider wants to offer broadband services but lacks a fixed infrastructure, or where a new service provider wants to enter a competitive market where DSL and/or cable are available.

Clearwire, Craig McCaw's latest venture in the wireless market, is an example of a new entrant marketing BWA broadband services in a few cities in the US where DSL and cable are available. It plans to start offering services later this year.

Fixed or mobile?

Initially, WiMAX will be mostly a fixed-wireless technology, and most CPE antennas will need to be mounted outside homes or office buildings. A version supporting mobility is due to be ratified by the WiMAX Forum in the first quarter of 2005, but Intel estimates that WiMAX mobile clients in laptops or PDAs will not be here for at least another couple of years. After mobility is introduced, connections can be maintained at speeds up to 150 kilometers per hour.

Mobility, however, requires more than a standard. Depending on spectrum availability, each service provider can deploy WiMAX using different profiles, either in the licensed or license-exempt spectrum. Each CPE may support one or more profiles, but in some countries, regulation may not allow a CPE to support certain profiles.

Subscribers will only be able to access networks that use the profiles supported by the CPE. In addition, mobile access will be available only where a service provider has deployed the network or has roaming agreements with local operators.

Support for mobility has an important consequence for BWA: its traditional focus on the under-served market could be greatly diminished. Top metropolitan areas will become key coverage targets, and BWA service providers will find themselves competing with DSL, cable and leased lines if they want to use the same infrastructure for fixed and mobile services.

The ability to share the same infrastructure has the potential to change the way broadband connectivity, both fixed and mobile, is used and marketed. Similar to the transition from fixed to mobile voice communications, data connectivity may cease to be linked to a physical location and become a more personal service. A move towards BWA from fixed and mobile service providers will result in bundled data services - including VoIP - that can be accessed regardless of location.

Service providers that currently depend on the incumbent infrastructure see BWA as an opportunity to switch to a facilities-based approach. In the US, for instance, unbundling of the local loop has come under regulatory pressure and may exist in the future in terms that are far less appealing to CLECs.

Covad in the US, a member of the WiMAX Forum, is evaluating BWA solutions. Allstream in Canada has started to deploy services, using its own backbone and working with network operators, service providers and in some areas AOL to provide service as a way to gain control of the last mile.

Mobile operators may find BWA attractive as a more cost-effective way to provide wireless data connectivity and as a way to transition to IP-based networks. That transition will require time, however. At first, mobile operators will see BWA technologies that support mobility as a competitor to their 3G data services.

In particular, they will be unlikely to commit to WiMAX while it predominantly addresses the fixed market. Taking a longer perspective, however, BWA will enable mobile operators to extend fixed-wireless substitution from voice services to data services, with wireless data connectivity sufficiently cheap and fast to replace fixed DSL or cable connections.

Background: WiMAX and 802.16

WiMAX is based on the IEEE 802.16 standard. But while 802.16 gives vendors a lot of latitude, and as a consequence does not guarantee interoperability among products from different vendors, WiMAX instead focuses on a limited subset of the standard. The initial three profiles are 2.5-2.7 gigahertz (licensed spectrum in the US market), 3.4-3.6 GHz (licensed outside the US); 5.75-5.85 GHz (license-exempt). WiMAX certification, when available at the beginning of 2005, will require products to interoperate with others operating within the same spectrum band.

Interoperability is a major advantage. It gives service providers independence from a single vendor and flexibility in designing a network. Perhaps even more importantly, the service provider is able to support any WiMAX client, and thus subscribers can select the CPE or mobile client they prefer.

WiMAX gives service providers the flexibility to use licensed or license-exempt spectrum. Established service providers have a clear preference for license spectrum deployments, as it offers better protection for their investment and they feel more comfortable guaranteeing SLAs to their business users. New entrants and small wireless ISPs are often more inclined to use unlicensed spectrum because it streamlines deployment and eliminates a big cost item from their business plans. A WiMAX base station can cover an area of up to a few kilometers radius and have 75 Mbps throughput, and residential subscribers can get a 1 Mbps connection or more, depending on their contract with their service provider. Security and quality of service capabilities are also built in to meet the requirements of business users.

Download article in PDF