Receive new posts as email.
This site operates as an independent editorial operation. Advertising, sponsorships, and other non-editorial materials represent the opinions and messages of their respective origins, and not of the site operator or JiWire, Inc.
Entire site and all contents except otherwise noted © Copyright 2001-2006 by Glenn Fleishman. Some images ©2006 Jupiterimages Corporation. All rights reserved. Please contact us for reprint rights. Linking is, of course, free and encouraged.
Part of the rollout of mobile WiMax requires good end-user products: Zyxel will provide PC Cards and CPEs (adapters for the home) that will be bundled with Nortel gear. Runcom is working on phone and laptop integration. Sequans is developing more spectrally efficient WiMax, and Nortel is working with them for interoperability testing of their MIMO-based WiMax products.
Senza Fili Consulting’s latest WiMax report predicts a relatively large market for both mobile and fixed uses of “mobile” WiMax: The report says that 54m subscribers could be signed up by 2012, but that emerging markets and the U.S. will be key to WiMax’s worldwide uptake. 54m subscribers will be a relatively low number compared to broadband—300m broadband subscriptions are in use today, according to a report earlier this week—and quite low compared to the billions of cell phone voice users expected by 2012.
The report author Monica Paolini notes that the truly mobile WiMax access devices—not nomadic ones that require AC power or are bulky—will lead to increased adoption. That’s the goal of Sprint and Clearwire, certainly; they want PC Card form factor WiMax cards next year. Paolini also suggests that portable data devices with WiMax built in will be key in developing nations for adoption.
The Wall St Journal reports that Sprint Nextel is looking at financing options for WiMax service: The company has committed $3b to the rollout, and may need to spinoff the WiMax unit and form a partnership with Clearwire. That would provide a massive amount of licensing power, and vastly reduce the competitive framework between the two operators. Sprint might also create a “more modest partnership” with Clearwire. Sprint might also look for outside financing, notably from MSOs (multiple system operators, or cable TV giants), with which they already have a relationship for reselling cellular service.
The article makes a few mistakes. While it notes that Clearwire currently uses a WiMax-like technology, the company’s path is for actual WiMax in 2008. That should have been noted, as it’s key in terms of integration between Sprint and Clearwire. Second, the chart showing a comparison of options between WiMax and 3G services has several problems. First, the bandwidth comparison doesn’t note that it’s downstream speeds only. Second, 3G networks using EVDO Rev. A, which is now extensively but not completely deployed by Verizon and Sprint, can reach 850 Kbps as a routine speed with peaks over 2 Mbps. The WiMax speed is noted as an absolute. Third, comparing 3G to DSL and WiMax to cable is absurd. DSL is routinely available at rates from 1.5 Mbps to 6 Mbps. Cable is typically only available at high starting points, but pricing is often in parity with similar DSL offerings.
Om Malik points out at GigaOm that the Sprint/Nextel merger may be the hidden cause of these plans, as the integration between the two firms has proven problematic. Nextel, however, had a large spectrum portfolio, and it was clear that was one of Sprint’s big motivations.
Horizon Wi-Com may be the first out of the gate due to more modest ambitions than Clearwire, Sprint Nextel: The firm has a 2.3 GHz spectrum portfolio that they told InformationWeek has allowed them to set up networks with Navini equipment in Boston, New York, Philadelphia, Washington, D.C., Baltimore, Pittsburgh, Buffalo, Richmond, and Cincinnati. Philadelphia is the only one of those cities with a citywide Wi-Fi rollout underway. Cincinnati has a smaller community-driven effort in place.
The company claims they’ll cover 70m people (POPs being defined incorrectly in the article). Pricing hasn’t been announced; a commercial rollout is three months away. They plan to clean Wi-Fi’s clock. Interestingly, the 2.3 GHz licenses were purchased from Verizon, InformationWeek reports.
The cities in question could likely see service also from Clearwire and Sprint Nextel in the 2.5 GHz band. Both firms claim to have licenses covering over 200m people, which likely includes these cities. So residents of some towns could see Wi-Fi, three competing mobile WiMax offerings, and three or four competing 3G cell networks (depending on when T-Mobile launches service).
The FCC, after nearly two years, has affirmed its plan to allow overlapping uses of the 3.650 GHz band: The agency has reserved 50 MHz for a special kind of licensed use that will make it possible to run reliable high-signal-strength systems from fixed base stations using WiMax or—an import or—any other standard or technology that conforms to what sound like fairly broad rules. The band will allow 20 watts of transmit power versus a maximum of 1 watt for an omnidirectional antenna with Wi-Fi and like systems in 2.4 GHz and some of 5 GHz.
The licensing require is just a bare minimum. There will be no restriction on the number of licensees, and licensees will be required to register themselves and how they’re using the band in their area. The users of the band will have to coordinate among themselves to avoid interference and system problems. Failing that, the FCC could step in. But no licensee under this regime will have any priority rights over any other.
It’s a significant departure from the Part 15 rules that govern unlicensed Wi-Fi and other devices in 2.4 GHz and 5 GHz, in which every device is required to accept interference without complaint, but not generate too much interference. The 2.4 GHz band in particular is a problem because there are many competing uses of the band, including licensed users that do have priority (amateur radio operators among others).
The band isn’t completely clear, however. Earth receiving stations for satellite systems will have priority over new uses, and they will have broad exclusion zones that will prevent this band from being broadly used in urban areas, especially on the coasts. Wireless ISPs and others who can, however, negotiate with earth station operators and come up with solutions that let them operate in the band.
Initially, therefore, rural areas will see much simpler use—with potentially no overlapping wireless ISPs, even—and, Harold Feld argues, so will municipalities. The band is perfect for backhaul and network links for a muni network, and cities are in good positions to negotiate with the existing licenseholders—they’re on the hook if something goes wrong.
The lower 25 MHz of the band will require scheduling of time slots among competing devices in the same airspace; the upper 25 MHz entire band will use open contention rules similar to those in Wi-Fi (listen to see if anyone’s talking, and then start talking if they’re not). That’s WiMax rules for the lower band, more or less, and Wi-Fi rules for the whole thing. (Note: I had originally stated that only the 25 MHz top half of the band would allow contention rules; not the case. See the comment from Harold Feld below.)
I’ve linked to Harold Feld’s analysis. Feld works for the Media Access Project, a non-profit that’s interested in expanding the number of voices that get heard by influencing media ownership, telecom, and spectrum policy. So he has a horse in this race. But his description is detailed and tells you who has won and who has lost (no one, really) in the FCC’s decision.
Meanwhile, Ubiquiti already has a piece of gear ready to go—it’s designed for either 3.5 GHz (a more broadly available worldwide chunk of spectrum) and 3.65 GHz with contention.
Om Malik argues that fixed wireless broadband provider Towerstream is riding Clearwire’s coattails: Towerstream has been providing fixed service for some time in major cities, but hasn’t had explosive success. They got into the over-the-counter stock market by buying an unrelated firm (University Girls Calendar Ltd.!), and then migrated into the NASDAQ. They’re looking to raise $40m, but at $4 a share—not yesterday’s close near $6 a share. Thus, a plunge. Revenues are quite low for such a long-established firm in mature cities.