Free Newsletter
Seybold: The New Clearwire- Will it succeed?
There has been a lot of positive press about the "new" Clearwire, a company that was put together from the original Clearwire, Sprint's Xohm, Google, Comcast, Intel and a few other players. This new company also received an infusion of $3.2 billion on top of what has already been spent by Clearwire and Sprint. The consensus in the press is that this new wireless broadband system will now have a chance to succeed.
Well, once again, I am one of the few who don't agree. Others point to the fact that the effort is led by Craig McCaw who has a "Midas" touch when it comes to wireless, that is unless you go back and look carefully at his track record and then you will find that perhaps it is not as good as myth has it.
But the real problem with the network is that those building it and those pouring money into it are hyping the capabilities of the network to the point where it will be a disappointment when it is turned on. WiMAX, as it is being deployed by Clearwire, is a 3G technology, not, as many people are saying, a 4G technology. The first true WiMAX mobile installation is being completed in Portland and Sprint has done a "soft launch" in Chicago, Baltimore and Washington, D.C. At this point, we have no published data from tests being conducted on the network. Nor do we have any results when it comes to speeds and network capacity. It is odd to me that we are still only hearing about the speeds that are predicted and not real-world speeds that have been measured in the field.
The big driver for this network is that it be what Google wanted from the 700 MHz auction--a completely open network, capable of being used for any information from any source. Any device capable of WiMAX can be purchased in a store and added to the network quickly and easily. There will be no walled garden. There will be nothing preventing you from getting to anything you want including Google searches, streaming audio and video, large files and more. Of course, Intel will be building a WiMAX chipset for inclusion in any notebook and any Internet appliance. In fact, Intel's vision extends to game devices, navigation systems and perhaps even items such as dog collars.
You will be able to buy a combination device that will enable you to use the Sprint CDMA network for voice and WiMAX for data. Or you will be able to buy a device that will enable you to use the Sprint EV-DO network for broadband data while the Clearwire network is being rolled out. When you are not in a city that has WiMAX, you will be able to use the existing Sprint high-speed network. My guess is that the existing Sprint network will have speeds and coverage that is so much better that people will wonder why they bothered with WiMAX. But that, too, will be due to the wave of hype in advance of the system.
We are being promised a fully open network, voice and data (VoIP on WiMAX), that will "deliver four times the throughput of other technologies at one-tenth the cost," according to Barry West, the then President of Sprint's Xohm 4G business unit that is now a part of Clearwire. Well, if you start adding up what Clearwire has spent plus what Sprint has spent, plus the new funding put into this new company, you end up with billions in costs. By the time they are done, they will be nowhere close to one-tenth of the cost of existing 3G systems. Consider this: To cover 75 percent of the U.S. population (not land mass) in the new 700 MHz band, you will need about 22,000 cell sites. To cover this same population at 2.5 GHz, which is where the Clearwire system is being built, you will need 65,000 cell sites.
WiMAX is not a bad technology. The current version has not been well designed for true mobility. It is a 3G wireless technology and it will be used in many places where there is no infrastructure available today. However, in an environment where there are three established broadband networks (soon to be four with T-Mobile joining the fray), I am not sure I would want to invest the billions it will take to build this system, especially when 4G technologies such as LTE and UMB are less than two years away from being deployed.
One final note here--no matter what is claimed, wireless bandwidth is shared bandwidth. A truly open network without any limitations as to who can download what and when could mean that some customers will have very slow connections while the kids next door stream their videos onto their devices. I wonder what those who believe Clearwire has to succeed because Craig McCaw, Sprint, Intel, and Goggle will make it succeed, will say when it doesn't.
Andrew Seybold is an authority on technology and trends shaping the world of wireless mobility. A respected analyst, consultant, commentator, author and active participant in industry trade organizations, his views have influenced strategies and shaped initiatives for telecom, mobile computing and wireless industry leaders worldwide. www.andrewseybold.com
Comments (18) | Post a comment
More stories about Walled Garden Sprint Google Comcast Clearwire Andrew Seybold 3G Xohm Wireless Broadband WiMAX
Comments
Posted by Anonymous | May 28, 2008 - 11:30am
WiMAX has a chance to be big in the US, but only if Clearwire can assume control from Sprint on how it's deployed and operated. Sprint has a huge mess on their hands with Xohm and the two other networks that have customers leaving by the millions. Yes, millions. They have proven they can't operate wireless networks wasting billions of investor’s dollars and with Xohm's constant embarrassing delays due to something as simple as backhaul, can't deploy a wireless network either. Clearwire and the other investors need to utilize Sprint's spectrum, give Sprint a seat or two on the board and then have their own management and engineering teams installed to run the show the day the merger is complete.
Posted by Anonymous | May 29, 2008 - 9:27am
Why does everyone see 700MHz as the Holy Grail? The operators have 15MHz each. Due to the FCC requirements of OOBE in adjacent spectrum and LTE channel profiles, the operators will be able to do a SINGLE 10MHz channel. Granted with MIMO technologies, throughputs could be high, BUT… N=1 reuse does not work well with OFDMA technologies. The sector overlaps will have very poor throughputs. Also, what happens if they exceed capacity for the sector, they have to build a new cell site. This will pose challenges as 700MHz carries so well. Did I also mention N=1 interference? The single 10MHz channel is very limiting when it comes to the services being pitched, such as IPTV and other video content.
Clearwire on the other hand has gobs of MHz in many major markets. They are looking to use MIMO as well, so throughputs will be good. In the future, they could go to more advanced MIMO such as 4x4 or higher to compete with LTE or even switch over to LTE. Either way, The ability to do N=3 or higher for reuse is a stark advantage. If Clearwire exceeds capacity, they add another channel, and still maintain N=3 or better. Interference management will be far easier and thus the system will perform better. Because of the MHz available, Clearwire will be able to offer many BW intensive services and will be able to service them for years to come.
Yes, 700MHz will cover for miles (a problem in the core), penetrates buildings better, and a host of other positives. That is great if you are covering rural America or if you do not need capacity. Thus with its ideal 2.5GHz spectrum Clearwire really has a chance to become the king of broadband. Assuming, of course, they can keep the direction focused and the partners happy.
Posted by Anonymous | May 29, 2008 - 9:14pm
WiMAX is an OFDMA-based, not CDMA-based, which means its a next generation, or 4th generation technology. In many cases, Clearwire will use six 10MHz channels. In a cellular configuration, that's three 10MHz channels and another three on top of that.
If mobile WiMAX was slower than 3G and had less capacity, do you really believe they would move forward with WiMAX? Don't you think they tested it?
UMB will be deployed? By who?
Posted by Anonymous | May 30, 2008 - 8:48am
UMB is offered by T-Mobile in the US. It is the WiFi at home.
Posted by Lynnette Luna | May 30, 2008 - 10:29am
T-Mobile using UMA. UMB stands for Ultra Mobile Broadband,a 4G standard designed for CDMA operators. No operator has publicly committed to UMB. Verizon has gone LTE as has Alltel.--Lynnette
Posted by Daniel | May 30, 2008 - 2:37pm
I think the best point you've made is the number of sites needed to deploy a 2.5 GHZ network vs. a 700 MHZ network. It's simple problem of physics and this problem will be much harder to deal with than people realize even after the network is deployed. They will just continue to densify and densify the network the traffic routing issues will be a nightmare if they don't define a deployment protocol from the start. And the old cellular protocols which is basically a manual design/deployment can't apply they basically need to invent a new protocol that focuses more on modeling and developing a good traffic routing architecture. Essentially, WiMAX is a FWA application that is trying to be used as a true mobile network and it may never really get to that point. Although is LTE a real 4G technology? I understand 4G to be a convergence of 3G technologies like WiMAX and LTE?
Regarding single frequency, isn't that what the whole subchannelization permutations (fractional frequency reuse) component of OFDMA technology addresses? Increasing bandwidth is not as big of a problem in theory according to the 802.16e standard. I think the first order problems will be dimensioning the network to start with at 2.5 GHZ (or worse 3.5 GHZ in Europe and South America). This physical layer modeling is what established operators could care less about and is what Andrew is trying to point out I believe.
Posted by Daniel | May 30, 2008 - 2:43pm
I meant to say physical layer design in my comment above, not physical layer modeling. Established operators don't have a clear protocol for designing a 2.5 GHZ network. You cannot assign 6,500 deployment/project managers to deploy a 65,000 site network. This 10 BS to project manager approach is what cellular operators are used to and at 2.5 GHZ it is too hard to work this way. You have to design the network with models much more accurately before going into any deployment.
The only operator that began to address this issue was Earthlink with their MuniWiFi business. Technically they were making advances but never had a solid business plan.
Posted by the Observer | June 1, 2008 - 6:35pm
As smart nodes continue to evolve, (think zigbee..) and femtocell type sites cheapen then wimax has a chance.
Couple it with the wifi infrastructure all ready deployed and there is a compelling business model here. And this is possible without just making it in the mobile voice market alone.
It does not need to make it work monitarily with just voice alone. This allows it to be deployed with some revenue until the voice capable network and phones are all there with full coverage.
The bigger question is will the other carriers be willing to play cutthroat with there pricing on data services ??? If so then clearewire will have a longer road to profitability. Given the players donot expect this to just falter if this turns out to be the case. It will just take longer to make a profit.
But a least us consumers will benifit by getting data services at a fair price vs now wher only must have bus types on the company expense can afford.
If phones are available that can go from wimax to cell seamlessly then wimax WILL flurish at the speed of light. (This may have to be done against the wishes of the established players, but hey a 13 year old can crack a phone faster than you can say wii!
Posted by observer | June 1, 2008 - 6:49pm
also, the reason the data services are so expensive is because the bandwidth is just not there on cell netwrks for all of us. that is why the price is so hi, If they offered a fair price for data now then your phone would drop calls all the time not just once every few days.
So the cell operators cannot play cutthrout with out doubling their cell coverage with new technologies and/or massive femtocell deployment.
now the interesting thing is sprint is part of this so it will get the benifit of backhaul/redundantcy for its network alone, alowing it to offer cheaper voice/data bundles over the cell only crowd. cheap all you can eat WIMAX and pay as you go eEVDO/LTE plus voice will be compelling for us penny pinching consumers if the right bundle is priced right.
Posted by observer | June 1, 2008 - 6:56pm
As a cell site coverage shrinks, by power or frequency limitations, the reuse gets better and better, the number of redundant available neighbor nodes increases, allowing for multiple paths to backhall locals. This reduces the bottleneck/and bandwidth of the backhaul node needed.
This continues to cheapen the site costs and the backhaul to make your cell site ratio point less and less relavent. Add smart (but need to be cheap) nodes and it may just disapear altogether.
Posted by Anonymous | June 1, 2008 - 11:53pm
Yes, as Prof. Willie Lu, Stanford Univ, already pointed out in 2006 that Sprint will be totally failed in this mobile WiMax movement. He listed 18 problems both in technology and business studies. See willielu.com
Posted by Anonymous | June 1, 2008 - 11:57pm
Wireless Communications Law:
No single wireless standard can do both broadband and seamless mbility. So theoretically, mobile WIMAX NEVER works.
4G is the converged wireless solutions based on open wireless architecture (OWA). WiMax is definitely NOT a 4G technology.
Posted by Amitabh Kumar | June 2, 2008 - 1:55am
It is not possible to agree with the contentions of future of the JV based on past performance of some people or technologies. 3G technologies, to me are those which make possible data links averaging 1-2 mbps while those of WiMAX which run easily at 4 Mbps+ in real life ( per user) fall in the next generation. 3G is focused on voice and TDM architectures, WiMAX is focused on QoS based service specific connections such as video or audio streaming. See the blog entry " I saw the future in may car" on Radio and Internet to just get a glimplse of what it can do. The partners such as Comcast and Time warner, Google etc have come onboard after the trials and not before. They must have seen something that we have not. More details
wimaxbook.net
Posted by Anonymous | June 2, 2008 - 4:23pm
I have to pull up Amitabh Kumar on his speed comments. I don't know where he got those numbers from but I can guarantee that "3G" technologies get a hell of a lot more than "1-2Mbps" in the real world. Down here in Australia we have a national (99% population) HSPA network and my real world sustained speeds in mobile environments are in the region of 4Mbps with peaks in the high 5Mbps range (this is with a 7.2/2.0Mbps device). This is plain old HSDPA/HSUPA. If this experience is anything to go by then I would expect throughputs in 2 figures to be quite easily achievable for HSPA+ and well into the 3 figures foir LTE.
Posted by Anonymous | June 2, 2008 - 4:29pm
in addition to the above, here is some evidence of the real world speeds (HSPA 7.2/2.0Mbps device on 14.4/2.0Mbps capable network:
www.speedtest.net/result/278987701.png
this was on the train on the way to work....
Posted by Anonymous | June 2, 2008 - 10:46pm
While it is true that LTE is 2 to 3 years away Qualcom canceled UMB a few weeks ago. WiMAX 802.16 2005 is mobile and availble with MiMO. And 700mhz does have great penetration and coverage but lousy capacity in metro areas. There is no free lunch. As to the 4g vs 3g show me a good definition. Is LTE better maybe but WiMAX will be on the next version by then and they will be about equal. Who ever has the ecosystem will be the winner.
