Telecom analysts seem in agreement that the dominant market forces in residential broadband Internet services will continue to be digital subscriber line (DSL) and cable modem technologies. Not just dominant: “all-powerful” would be a more accurate description. Cable modem and DSL have become the default technologies for consumers who desire to make the move up from dial-up service. Although there exists some dispute regarding which of these forces will ultimately come out on top, most analysts agree that alternative Internet service providers such as fixed wireless or satellite-based broadband will at best only collectively occupy about 8% of the total broadband Internet market in 2004.

The most commonly stated reasons for this widespread belief can be summarized as follows:

1. The technologies are unproven on a large scale.
2. The market is too stratified to admit another major technological competitor.
3. Customers are not familiar with these technologies.
4. Such technologies are too expensive to implement.
5. There are problems inherent to these technologies that limit their wide-scale use.

AVToolBox Home Theater Components
Although this list is not exhaustive, it outlines the most potent arguments against aggressive, short-term expansion of alternative residential broadband Internet services. Although I tend to agree that it may be a few years before these technologies will effectively challenge cable or DSL for market dominance, I believe that a technology exists today that will penetrate the broadband Internet market both sooner and more profoundly than most market analysts suggest. That technology, in my opinion, is MMDS.

Overview of MMDS

Multipoint Multichannel Distribution Service (MMDS) is not a new technology; it has been used to transmit wireless analog cable television signals for more than 30 years. It is only recently, however, that MMDS has been digitized and adapted to transmit data and Internet traffic.

The FCC had originally allocated spectrum in the 2.15 – 2.16 and 2.50 – 2.686 GHz bands for broadcast services capable of providing 33 traditional 6 MHz video channels. Twenty of these channels were set aside for Instruction Television Fixed Service (ITFS), and thirteen for Multichannel Distribution Service (MDS) and MMDS. The ITFS channels were originally intended to be used by non-profit entities such governments and educational institutions.

In the beginning, ITFS spectrum holders found it difficult to financially support full-time use of the allocated spectrum, so they leased the excess space to cable operators. Many cable companies took advantage of this additional capacity to expand their services to those markets without more conventional access to CATV.

Only recently, however, has commercial interest in leasing the ITFS spectrums become noticeable. Once the FCC expanded it’s definition of “educational usage” on the ITFS band to include both voice and data services, and reduced educational programming requirements to 20 hours per channel per week, many cable operators and other telecom companies became curious. Perhaps the financially strapped wireless cable industry finally may have finally discovered its “killer application”: wireless Internet services.

In MMDS they discovered a technology that can be used to deliver multiple channels of video, data and voice signals (typically in a 25 to 30-mile radius), and can offer high-speed Internet access (from several hundred kbps to 10 Mbps) using only slightly modified versions of cable modems. MMDS offers two-way potential, and through the utilization of various frequency reuse methods, including antenna sectoring and cellurization, a single channel can support hundreds of users simultaneously. Moreover, through the use of superchannelization (i.e., the aggregation of multiple contiguous channels of standard bandwidth into channels of larger bandwidth), MMDS providers can choose the bandwidths they use at their stations, and offer adjustable bandwidth according to specific customer needs (bandwidth-on-demand).

Overcoming the Objections

Why are wireless ISPs, including MMDS-based services, anticipated to at best enlist less than 8% of total broadband subscribers in 2004? Several arguments have been offered against MMDS technology, in particular, but we will be best served by limiting our attention to the five mentioned previously. They are, after all, the arguments most frequently offered by broadband pundits.

1. The technology remains unproven on a large scale.

As previously mentioned, MMDS technology has been in service for about 30 years. There exists a wealth of experience testifying to the validity of MMDS technology, albeit primarily one-way analog video. But the jump to two-way digital video, voice and data transmission has been relatively effortless, requiring little more than regulatory approval. The real essence of this argument seems to be a concern over the scale of deployment.

No one will deny that there is limited empirical data regarding two-way MMDS services. Since two-way MMDS was not approved by the FCC until 1998, the landscape is sparsely populated by less than twenty MMDS ISPs, most of them offering only one-way MMDS service, with regular phone lines being used for upstream signals. That is, until recently.

Several beta trails using two-way MMDS technology for high-speed Internet services have been initiated, and in many cases concluded, with great success. Sprint’s recent trials in Tucson were so successful that it has initiated MMDS Internet service in Phoenix, with combined service supplied to more than 8000 customers. Hoping to build on this success, Sprint plans to launch MMDS service in fifteen to twenty metropolitan areas by year end 2000, and hopes to fully utilize its MMDS spectrum licenses in 90 additional markets within the next two years.

Despite the failed WorldCom/Sprint merger, WorldCom continues to move aggressively on the MMDS front as well. WorldCom currently owns MMDS spectrum in 160 markets, and is conducted trials in Jackson, Mississippi, Baton Rouge, and Memphis, and will soon begin trials in Boston and Dallas. Memphis will go live with commercial service by the fourth quarter of this year, with other markets expected to follow soon after. “Phase One” of WorldCom’s national MMDS plan calls for service to be offered in 60 markets by year end 2001, and 100 by year end 2002.

Unproven: maybe. Unrealizable: tell that to WorldCom and Sprint. Certainly commercial rollouts of this magnitude convey unquestionable technological credibility for MMDS.

2. The broadband ISP market is too stratified for MMDS to constitute a significant threat to cable or DSL.

Cable and DSL certainly are the dominant forces in today’s broadband Internet market. Projections, including Parks Associates’ very own, predict that DSL and cable broadband will remain dominant for many years to come. But is this market too competitively stratified for an alternative broadband technology such as MMDS to threaten the dominance of these two beasts of broadband?

Answering this question, for one, requires an understanding of the nature of the broadband Internet market at current. The competition between DSL and cable for the heart and soul of the broadband Internet market is vicious, even downright nasty. In the most recent phase of the battle, DSL is targeting the shared bandwidth of cable Internet as a possible cause of civil unrest between neighboring suburbanites (even though DSL itself requires the sharing of bandwidth, albeit further up the pipe). In other headlines, the telcos are attempting to achieve a tentative parity in cost of service by undercutting the prices of cable Internet service. No doubt this has placed pressure upon the already razor-thin margins of most cable ISPs, as SBC has recently complained. Adding a third component to this struggle, especially one that more than likely will be courted by the existing cable companies, will further charge the competitive atmosphere.

Secondly, it has recently been suggested that wire-based ISPs may in fact be approaching a point of diminishing returns. Last mile strategies are becoming increasingly difficult and expensive to undertake, for both cable and DSL companies. Easy-in, easy-out service deployments are all but exhausted, leaving the more labor- and time-intensive markets to be developed. At what point does cost and time-to-market considerations favor a new alternative?

Third, despite the belief that cable and DSL compete head-to-head in most areas, the vast majority of customers do not have a real choice between one or the other. Rather, the choice is more likely to be between dial-up services, on one hand, and cable or DSL, on the other. With the exception of most suburban areas, the greater part of broadband Internet market remains without a true choice of options. If MMDS can serve just a portion of those without Internet service, or those Internet users who have yet to upgrade to broadband services (a market which, by the way, is said to be somewhere in the range of 70 million customers), then it will not take long before MMDS can compete head-to-head with the subscriber base of both cable and DSL.

Fourth, and perhaps the most important reason why MMDS may indeed arise as a worthy technological antagonist to cable and DSL, is the deep pockets and reputation of those companies whom are playing the MMDS card: Sprint and WorldCom, in particular. In 1999, Sprint and WorldCom combined spent an estimated $2 billion in acquiring analog wireless cable TV service providers. Why? Because these small cable companies owned the MMDS spectrum in highly prized areas. Sprint and WorldCom purchased the debt and equity of these analog companies because the frequencies they held could now be utilized to supply digital two-way services over the “last mile” of Internet access. It is critical to understand that these frequencies are issued by the FCC for each specific market, and are in many cases licensed to the owner as a protected market area known as a “Protected Service Area” or PSA. In other words, no other entity can compete with the frequency owner in its PSA.

With rights to MMDS spectrum in over 300 markets, Sprint and WorldCom’s combined footprint would exceed 54 million people (Parks Associates’ estimate). By any calculations, that constitutes significant market potential. With a market this size, and with exclusive frequency rights in many areas, both Sprint and WorldCom are poised to alter the landscape of broadband residential Internet service.

3. Customers are not familiar with MMDS technology, and this will hinder widespread utilization of MMDS.

True, but certainly this concern is not unique to MMDS technology. This is a sociological phenomenon that is well documented. Any new technology, cable and DSL included, must at some time deal with the cultural lag created with the perpetual introduction of new forms of technology.

It was 1923 when University of Chicago professor William Ogburn introduced the concept of “cultural lag.” Ogburn posited that as new forms of increasingly sophisticated technology are introduced within a culture, parts of that culture will react in strikingly different ways, and at different speeds. New technologies, according to Ogburn, tend to be adopted if and only if they can be proven to have “superior utility.” Simply stated, as individuals actually see how a particular technology functions, they will evaluate its applicability to their own needs and make a decision as to whether or not to adopt the new technology.

One doesn’t have to remember too far back to recall a time when cable modems and DSL were foreign technology to the American consumer. And although in many ways cable and DSL are still fighting hard to overcome the natural tendency of consumers to resist new technologies, slowly and surely the market for their services has grown and will continue to grow as subscribers become more demanding of dial-up services and come to realize the advantages of adopting new technologies.

4. MMDS technology is too expensive for mass deployment.

This concern is misguided at best. WorldCom’s internal estimates are that to build-out new MMDS transmit/receive sites costs about $2000/square mile, along with a minimal investment by the customer for a small external antenna and a modem wired to the antenna (which is in many cases subsidized by the provider). And once service is rolled out, it is estimated that customers should be able to receive service in a week to ten days. This is substantially less expensive than new DSL or cable lines, and cheaper than other alternative broadband Internet technologies, such as local multipoint distribution systems (LMDS) or satellite-based broadband service.

In fact, MMDS’s unique ability for mass deployment renders the technology even more cost effective than other alternatives. It is precisely because MMDS service does not require as many transmitting towers as LMDS, serving thousands of subscribers from a single transmitter, that Sprint and WorldCom have chosen MMDS as the alternative broadband Internet technology of the future.

In many cases, transmitter construction will actually be unnecessary, which will further reduce initial deployment costs. For example, Sprint has many towers already built, and is uniquely positioned to take advantage of existing network agreements encompassing over 15,000 transmission towers throughout the country.

5. There are problems inherent to MMDS technology that limit its wide-scale use.

No communications technology is immune to difficulties. MMDS is a microwave-based technology, and as such has to contend with its own unique set of problems. The two most common concerns deal with (1) the distortion and interference inherent in multipath transmissions, and (2) line-of-sight requirements.

Although it is true that many microwave-based communications networks are susceptible to interference caused by weather, MMDS is immune to the most common weather-related culprit: rain fade. At high-frequency bands such as 24, 28, and 38 GHz, wavelengths are short enough that raindrops can actually present line-of-sight obstacles and greatly attenuate the signal. With MMDS, because of its longer wavelength, rain represents no obstacle. There are years of continuous transmission history that validate this fact.

The multipath distortion limitations of MMDS are well understood, and many companies have introduced products to deal with precisely these concerns. For example, Gigabit Wireless is developing a “smart” antenna that helps to eliminate both multipath distortion and interference issues, as well as line-of-sight problems. Commercial availability of this antenna will be year-end 2000.

Moreover, with the introduction of the Wireless DSL Consortium in July 2000, many of the industry’s most prominent equipment manufacturers (among them ADC Telecommunications, Gigabit Wireless, Conexant Systems, Intel, Nortel Networks and Vyyo) came together to begin the task of constructing a common set of open interfaces for MMDS services. Such a common standard would go along way towards reducing interference between MMDS license areas.

In terms of line-of-sight requirements, with the introduction of new technologies (as previously mentioned) and multiplication of line of sight opportunities through cellurization, although such obstacles cannot be completely eliminated, their effect upon transmission signals can be rendered functionally insignificant.


Given the perpetual introduction of new broadband Internet technologies, the commercial dominance of cable and DSL will undoubtedly become less secure. However, most analysts project that the extent and velocity of market penetration for these alternative technologies will be minimal in the next four years, constituting no more than 8% of the total broadband ISP market in 2004.

The justification for this consensus has been outlined above. These judgments, however, are at best erroneous when applied to multichannel multipoint distribution system technology, or MMDS. The two-way digital capability of MMDS is clearly established; deploying MMDS service is less expensive than alternatives; the problems inherent to MMDS transmissions are being minimized; and with the deep pockets and aggressive plans of WorldCom and Sprint behind it, MMDS is likely to become a legitimate, commercially viable broadband Internet technology in the next few years.

MMDS is ideally suited for areas where DSL and cable cannot reach. And in areas where these services are available, MMDS can offer ISPs the advantages of rapid deployment, a swift time-to-market, none of the “last-mile” obstructions inherent to DSL and cable service, and a more cost-effective mode of providing Internet service. All of these advantages should prove very enticing to ISPs scrambling to establish a presence in the broadband market. As such, MMDS will prove to be a worthy competitor for both cable and DSL, and the attention of Internet subscribers.