From Andy Seybold:
Sprint Gives WiMAX a Real Boost
The WiMAX community, powered by Intel, desperately needed a major network operator to commit to deploying WiMAX. Even with a $600 million infusion from Intel, Clearwire wasn't enough to make a splash because it's a portable-data-only network.
Sprint is another story. It has two fully mobile wireless networks up and running, one based on CDMA2000 including 1xEV-DO, and the other based on Nextel's iDEN technology.
Sprint is also involved in a joint venture with a number of cable companies that intend to offer their customers wireless options including voice and data with full mobility and integration with Wi-Fi for better in-building service. With the WiMAX announcement, these companies will be entering the high-speed wireless Internet game in competition with wired DSL and wired cable company data services. The cable companies recently indicated they might go mobile with WiMAX, so they will have the Sprint EV-DO Rev A network and its WiMAX footprint, currently targeting about one-third of the U.S. population. All of these companies could be competing with each other.
The WiMAX community desperately needed this first real win for WiMAX and the WiMAX community is probably celebrating as you read this. But let's look at exactly what Sprint has where, how WiMAX stacks up and what I think we can expect from this announcement.
First Things First
WiMAX is NOT a 4G technology. In fact, if you search for a definition of 4G using Google, you won't find one. The ITU defined 3G, but its site provides several presentations from various organizations and companies attempting to define 4G. I found no clear, concise definition. Most discussions about 4G talk about a full IP system with Quality of Service and DSL-type speeds. I guess the WiMAX community thinks WiMAX is 4G based on these criteria. However, if you base the definition on performance, it is clear that WiMAX is only on a par with HSDPA and EV-DO Rev A when it comes to bits per Hz. With IMS, there will be little difference.
Because these technologies use different bandwidths, it is difficult to do a head-to-head comparison. When we mapped WiMAX, UMTS/HSDPA and EV-DO Rev A (as well as other technologies) using a 10-MHz channel allocation, we found that theoretical peak data speeds of WiMAX exceeded those of the other technologies. However, when we normalized for typical data speeds, the three were close to a dead heat and I dispute the claim by Sprint CEO Gary Forsee that this fourth-generation network would be about four times faster than current high-speed cellular data networks. If that is the case, why doesn't Sprint put WiMAX in at 1900 MHz to cover its entire footprint instead of on 2.5 GHz where it will only cover about 100 million potential customers by the end of 2008?
The WiMAX folks will tell you that adding MIMO (multiple antennas) to their system gives them an advantage, but both UMTS and EV-DO have MIMO components in their next releases so the comparisons will remain the same: essentially dead even. (Just a note, tower owners charge their customers per antenna, so adding more antennas drives up the opex.)
All things considered, I, for one, do not recognize WiMAX as a 4G technology. It is another interesting technology that competes with EV-DO and UMTS/HSDPA that, at the moment, does not have a voice component. Sprint made it clear that this WiMAX network is not simply about notebooks and DSL/cable-like services but is designed to provide DSL-like access to all manner of consumer electronics devices. I am sure that if Sprint adds Voice Over IP services to its WiMAX system that is being designed for portable or fixed use, it will end up with two networks: one for fixed use and one for full mobility. Adding an IMS back-end and some Wi-Fi will result in a fairly complete set of offerings for home and business that will include full mobility voice and data and fixed voice and data via WiMAX and Wi-Fi.
Coverage
The next issue is how much of the nation Sprint can cover with its existing 2.5-GHz licensed spectrum. The company says that since the merger with Nextel it can provide service to 83 of the top 100 U.S. markets.
Sprint's CDMA coverage is much better than the 100 million pops it will be able to cover with WiMAX by the end of 2008 (unless it ends up with some AWS spectrum and decides to use that for WiMAX). Sprint claims that when it comes to coverage, the WiMAX network will be less expensive to build than CDMA or UMTS networks. How can that be? To provide the same coverage as its CDMA network, Sprint will need about twice the number of cell sites (although as a bonus, capacity will be increased with more sites). To match the footprint of its iDEN system or Verizon or Cingular's 800-MHz systems will require four times a many sites at a capex increase of more than 300%.
Costs and Rewards
The costs that were made public include between $2.5 and $3 billion over the next two years to cover 100 million pops that are, for the most part, already covered by wireless networks, muni-Wi-Fi, wired and cable providers and perhaps a few other competitors after the AWS auctions.
It is not clear how much Sprint's partners are putting up to get this system up and running. Sprint only alluded to the fact that its partners would be providing some funding. What do the partners have to gain? According the joint press release, Motorola will sell its single (WiMAX-only) and dual-mode (WiMAX/EV-DO Rev A?) devices and base stations to Sprint, Samsung will provide the back-end, cellular phones and other devices, and Intel will provide the WiMAX chipsets. I wonder how many chips Intel will have to sell to make up the $600 million it invested in Clearwire plus whatever it spent to help Sprint make up its mind. Intel's track record for follow-through isn't great. It was the first U.S. promoter of Bluetooth, making a big deal of it and then never incorporating it into its own chipsets. Then there was Cometa, the failed joint venture with AT&T and others, to place Wi-Fi hotspots within a five minute walk or drive from each citizen in the United States.
Intel will probably continue to throw money at WiMAX all around the world, hoping enough notebooks and other devices with WiMAX chips inside will be sold to recoup its investment and that all the network operators, including Sprint, realize a decent return on their investment.
Getting WiMAX to Market
Now the real work begins: installing the network, readying devices for market and determining real-world data speeds and how much capacity per cell sector to build out. (Like all wireless and some wired technologies, the total sector capacity is shared among the total number of customers.)
Sprint says it chose WiMAX because of the large number of vendors that are in the market or are coming to the market with product, and because some competitors are currently single-source suppliers. This will turn out to be the right decision if WiMAX lives up to its hype. Yes, trials are underway. Motorola, among others, has a number in progress. However, this will be the first full-scale deployment and, as with any new technology, there will be lessons to be learned, tweaking to be done and compatibility issues and other glitches to be resolved.
It will be interesting to watch how WiMAX fairs on 2.5 GHz competing head-on with full mobility players and their one-way TV systems (MediaFLO and DVB-H), the muni-Wi-Fi crowd, wired and cable companies and any new companies that might acquire AWS licenses. With LTE and 802.20 coming by 2010 or so, another question for Sprint and the WiMAX community in general is whether they can deploy and earn a return an investment in two to three years? Is there a demand for more broadband services where there are already services today?
If WiMAX was, in fact, four times faster than existing technologies, I would say it has a chance. But I don't believe these claims and our research shows that WiMAX, EV-DO Rev A, and UMTS/HSDPA/HSUPA are dead even when it comes to data speeds and capacity. If WiMAX was a true 4G technology, I would be excited about its chances for success. But it is not. It stacks up nicely against the incumbents, but it will eventually come down to device volume and pricing.
I have heard that the WiMAX community claims there is a cost savings on a per-cell-site basis and perhaps this is true; we won't know until we see a full commercial roll-out. I believe that at the end of the day the total system cost will be about the same given the difference in spectrum bands and number of sites that will need to be built.
Conclusions
Will Sprint be successful with WiMAX? I have said for a long time that there is no way that a standalone, data-only system (even with VoIP) can make money. But Sprint is adding another network into its system and will be sharing customer support facilities, back-end services, tower locations and other infrastructure that will enable it to bring WiMAX to market more quickly than a new operator. It will be able to bundle WiMAX with other services and, at some point, offer devices that will work over both its CDMA network and WiMAX network if there is sufficient demand. If anyone can make a go of WiMAX, it is a company such as Sprint with other revenue streams to support its business. It won't be long before we know if this new wireless technology, brought to us by the computer industry, can compete with technologies that have been brought to us by the wireless industry and new technologies that are only two or three years away.
What would I have done if I had been Sprint? Good question. Let me see if I can answer it. I probably would have given WiMAX a shot, but I don't think I would be standing up on a stage promising things my vendors tell me they can deliver in a commercial system. It is always better to promise less and over deliver than to over promise and have to make excuses later. And I have to think I would be wondering if two to three years is sufficient to earn a payback on my investment. One more thing, I would have to wonder how my business model would be impacted if a company claimed rights to some intellectual property and asked for royalties, even though I had been assured that would not happen, and now that Sprint and Clearwire are both using WiMAX, I would wonder about the disposition of the lawsuits that are flying back and forth. And I would have to wonder how my business model would be impacted when the 700-MHz actions are concluded in a few years.
Now that Sprint has announced its intentions, other major wireless providers can be expected do follow and we will see if the WiMAX community can really deliver what it has been promising or if WiMAX is simply one more 3G technology!
Andrew M. Seybold
--------------------------------------------------------------------------------
The Outlook 4Mobility provides its commentaries free of charge. Outlook 4Mobility products and services include Consulting Services, Mobiltorials, Newsletters, Customized Proprietary Research, Wireless Tutorials and Conferences. Please visit our web site at www.4mobility.com for additional information.
Copyright 2006 Outlook 4Mobility |
