Qualcomm gets cozy with LTE, makes migrating from CDMA a snap

What if Toshiba were to produce a Blu-ray player? If there's one surefire sign that a company is recognizing the mortality of its own standards, it's throwing some support behind the competition's -- and that's exactly what Qualcomm has done in announcing new roadmaps for its mobile and cellular base station chipsets that include LTE. LTE, one of several 4G standards competing for the hearts and minds of carriers across the world, has a huge leg up on Qualcomm's own UMB and WiMAX (which is technically a pre-4G standard, anyway) by having the blessing of the GSM Association, the global juggernaut of mobile industry organizations. Anyway, Qualcomm's new plans call for future chipsets to support various flavors of UMTS, HSPA, and EV-DO, theoretically making it easier for carriers of all creeds to migrate to LTE while still supporting legacy cells and devices. The new silicon is expected to be available next year, and without a single major carrier having signed up for UMB, we'd say that's not a moment too soon.

[CB17 @ Feb 8th 2008 4:52AM]

I almost feel bad for those wasting their money on WiMAX.

Nahhhh....

[CB17 @ Feb 8th 2008 4:24PM]

@wfactor9

Wow really? They're positioned nicely cause of the BANDS that they use? Do you understand ANYTHING about how cell phones, or just technology in general, works?

[CB17 @ Feb 8th 2008 4:25PM]

Damn commenting system, that was a response to below.

[Darkest Daze @ Feb 8th 2008 5:11AM]

So, are we one step closer to a world where I can buy any phone and use it on any carrier? Someone tell me we are.

[CB17 @ Feb 8th 2008 5:26AM]

Sure, as long as you use a GSM carrier. That's not new. There's only a handful of carriers in the world you can't do that with.

[Darkest Daze @ Feb 8th 2008 5:54AM]

Well since Verizon and AT&T will be moving to LTE, I think that 90% of people will be GSM once this next generation of mobile phones takes hold of the market.

[CB17 @ Feb 8th 2008 6:11AM]

Click the read link in the article.

3 different chips for the different carriers. It's too expensive for ANY carrier to try and redo their whole network. For Verizon LTE will simply sit on top of their existing CDMA stuff. At least for the next few years anyways (I'd say 3-5 minimum).

[Rich @ Feb 8th 2008 8:07AM]

LTE follows the GSM philosophy of supporting both voice and data. In theory, Verizon should one day be able to switch off their CDMA network. Having said that, Europe has been using UMTS for 5 years now and there's no sign of the old GSM networks being switched off.

Let's also hope that all the countries manage to find a common frequency to use...

[wfactor9 @ Feb 8th 2008 12:59PM]

@Rich

That's where Sprint is nicely positioned...People usually view WiMAX / Sprint as a technology choice...What they overlook is the spectrum issue which you, cleverly, bring up:
a) East / West Europe is deploying 3G in the 2500 bands (news flash: that's Sprint's bands in the US
b) UK is beginning the auction process for 2500 both TDD and FDD (translation: WiMAX in 2500 in UK, LTE in 2500 also)
c) Asia and India are already deploying 2500 WiMAX (and 3500 also)

Sprint's bands are well positioned...Most in the 4G circles believe WiMAX / LTE dual mode devices will be available - to account for WiMAX deployments today and LTE forecasted deployments in the future.

[Otto Khera @ Feb 19th 2008 2:02PM]

Question to Evengrift or someone else who might know: Can LTE operate in the 2.5 GHz band? Have the Sprint trials to migrate from WiMAX to LTE in the 2.5 GHz proved successful? Is this Qualcomm's game then, to create chipsets with different standards embedded, including CDMA (EVDO), WiMAX, LTE, and perhaps WCDMA?

[Kriston @ Feb 8th 2008 2:46PM]

That's because UMTS dovetails nicely off existing GSM networks and co-exists well.

[AndrewNeo @ Feb 8th 2008 9:09AM]

As long as Qualcomm stops doing BREW on LTE devices (or Verizon really does let you use your own phone)

[absurdio @ Feb 8th 2008 11:29AM]

Day after day, I wonder: Why did I pick Verizon?

[Frankenstein Black @ Feb 8th 2008 2:50PM]

R.I.P. (CDMA, EVDO, UMB & WiMAX ;^)...

[evengrift @ Feb 8th 2008 2:50PM]

I'm calling bullshit here, WiMax is to be a 4G Standard by any
measure I've seen out there. The standards for the bandwidth rates
need to be written, but no one else has written such standards
either. Keep in mind the WiMax standard in place now is 802.16e-2005,
as in as of 2005, lots of overhead here folks, they'll reach the bar
for 4G without a problem. Only the GSM loons would argue otherwise,
because tdma is just *such* a good technology. And hard hand-offs are
still in effect for the vast majority of carriers.

Calling 802.16 not a 4G standard is like saying 802.11 = 802.11b.
Anyone still selling 802.11b routers?

You do not improve throughput by slicing allocations by time and
frequency. You have to move voice out of that sillyness to free-up
that frequency for technology.

Lets take a dose of reality folks:

GSM now embraces CDMA (with the WCDMA derivative), a qualcomm/MOT
technology. GSM with LTE will move further into the CDMA space, as
leaders (WiMax/HIPERMAN/WiBro), move on to OFDM.

VZW is a wireless carrier sourcing the majority of their hardware
with Qualcomm chipsets. VZW has announced their intention to move to
LTE, obviously in part to enhance international roaming. Who doesn't
think Qualcomm wants in on that action?

Now lets look at LTE deployment vs. Other Technologies.. How much LTE
has been built? O, zip, nada, nothing. Designed? Uhhh.. yeah, same
zip, nada, none, at least they've ratified a standard... Uhh about
that, "we're working on it". Great.

What does this mean for a commitment from Qualcomm then? Absolutely
nothing, nada, zip, zero, this costs Qualcomm nothing, and they get
to blow the VZW a little public kiss.

And before anyone has delusions of HSOPA grandeur, when do you think
thats going to happen? consider the slow embrace of WCDMA (Hello
T-Mobile, where art thou?). And where are they going to slot it? Not
compatibe with CDMA derived technologies, or TDMA for that matter.
Anyone think ATT is going to start pulling down the HSDPA/HSUPA
equipment they just started to put up?

[CB17 @ Feb 8th 2008 4:21PM]

Wow so many things wrong where to begin...I think I'll just make bullet points.

-WCDMA has nothing to do with CDMA other than the name (not entirely true but mostly true
-"GSM with LTE will move further into the CDMA space..." Okay this line doesn't even deserve a response.
-T-Mobile hasn't launched 3G only because of the government.
-I'd rather wait an extra year for LTE which will provide 6x the bandwidth of WiMAX and also provide global compatibility
-I think you work for Qualcomm or Verizon

[evengrift @ Feb 8th 2008 4:48PM]

@CB17,

I work for neither, and wouldn't be calling bullshit if I did, would I? Make some sense here.

W-CDMA = WideBand CDMA, Which is a derivative of CDMA Used by the GSM Folks (among others). It is not compatible with QAM, OFDM, or TDMA, it can not share frequency with them. Nothing to do with CDMA? Are you completely out of touch? Who produces WCDMA chip sets? Qualcomm.

GSM has two choices: open up another incompatible technology (OFDM possibly as in HSOPA) and need more frequency, or continue with CDMA. QAM has inherent limitations and can only be pushed so hard here. Like it or not you may be an LTE Fanboi, but it isn't going to be magic that makes it work.

T-Mobile's problem has nothing to do with the government, they blew it because the negotiated a bad deal from a provider that gave them access to existing an 3G network, but didn't require the provider to vacate for another 2 years. Worse, they're also trying to get out of auction non-standard 3G Bandwidth, http://www.engadgetmobile.com/2006/10/06/t-mobile-details-3g-plans/ when they get it, they'll then have to source yet-to-be-built equipment.

Wait all you want for LTE, if it ever gets here, it'll still be behind 802.16(x). 802.16 is at the beginning of its life, NOT the end, or are you still using 802.11b?

GSM apologists all seem to be the same "I'll wait, our future is better." But *where* is this future?

Lets not forget the dirty little QPSK secret of HSDPA/HSUPA unless you've got a clean signal, you're not even gonna get QAM level efficency. Unlike EVDO, which while having other limitations doesn't fall back to such inefficiency.

802.16e will push OFDM forward, escaping the EVDO/HSDPA/HSUPA QAM badness, at least the EVDO vendors for the most part aren't still dealing with bandwidth draining TDMA customers.

[CB17 @ Feb 8th 2008 10:43PM]

No, you don't get any credit for copying the first few lines of wikipedia. And if you insist you should at least include the whole piece which talks about how W-CDMA is just the AIR INTERFACE for UMTS which still uses the GSM infrastructure.

Notice in my original post I said "not entirely true". This is because simply saying that those 2 technologies are the same solely because of the way they differentiate subscribers is ridiculous. Saying that CDMA and W-CDMA are the same thing is like saying that apples and oranges are the same thing just cause they're both fruit and both round.

http://en.wikipedia.org/wiki/UMTS

If it can't share frequency then how is it possible that ATT uses 850 and 1900 for HSPA as well as its original GSM technology?

"If LTE ever gets here" You're not really serious right?

As for being a fanboy, I'm simply a fanboy of the better technology and numbers don't lie. Your only rebuttal for LTE is it's "not here yet." Hey well guess what NEITHER IS WIMAX (I mean, I guess it's sorta here but it really isn't).

As for comparing WiMAX to 802.11b, if you're implying that the 802.16 standard is UPGRADABLE then, sure. But what's the point if it requires all new base stations and network cards (b -> g -> n)

I like how you totally made some random irrelevant negative remark against HSPA but totally glaze over EVDO as just having "other issues".

I guess defending WiMAX and EVDO as much as you do you must work for Sprint, then (or one of its partners). If you're gonna talk about something you should put a disclaimer down like Engadget does talking about AOL and Time Warner. Here I'll start, I work for a company that builds cell towers, we're currently working with Sprint on it's WiMAX rollout (It's a private company so I'm not gonna give away it's name online for obvious reasons). See, that easy, you try.

As for T-Mobile, I'm not even gonna begin to try and analyze THAT situation. All that's important for you to get through your "fanboy?" brain is that they're lagging behind has absolutely no bearing whatsoever on LTE's development.

[CB17 @ Feb 8th 2008 10:44PM]

See I can make super long comments, too. Isn't this fun?

[evengrift @ Feb 17th 2008 12:06PM]

I copied nothing from wikipedia. You're an idiot to say "just an air interface," your attempting to trivialize the most significant part of the equation. You have to multiplex the available frequency somehow to get the most out of it. GSM by in large is still using really really bad time division multiple access, aka TDMA, to accomplish this. Where they aren't, they have moved to CODE DIVISION MULTIPLE ACCESS (W/CDMA) to multiplex the service. And for data, limited to HSDPA/HSUPA they're using AMPLITUDE MODULATION, in the form of QAM. QAM is bad in mobile applications because impedance is not fixed and can only be slightly controlled for.

I do not work for Sprint, or any Telecom, a supplier, or even in a related field. I do however work with logic, theory, and ecc's.

ATT is operating HSDPA/HSUPA at 2100Mhz here in the states with little exception for where they've truncated older bandwidth to make room. It is not co-existing in the same time/place as TDMA multiplexed signals, as QAM/QPSK is incompatible.

I am serious about LTE never getting here, at least in its current form (which is nearly non-existent) or with its current members (seems to be everyone, with no commitment from anyone). Look at the slow-up

What irrelevant remark? Your irrelevant claim to an irrelevant remark? EVDO issues as far as I am concerned are bad, actually the only things I have to say that are truly positive about EVDO is that it works reliably, its widespread, and its cheap. EVDV was a better technology, but wasn't ready in time (was ready in time for GSM carriers, but they're too biased against technology coming out of CDMA schools). EVDO shortcomings imo are: no native support for voice, single channel (currently), and use of QAM.

How does one of GSM's larger members inability to adopt current standards NOT cast doubt on the embrace of future standards?

[misch @ Apr 15th 2008 6:20PM]

Thank you for writing a very funny comment. You seem to be confused about the technologies used, that makes this way more fun to read :).

[JimJam @ Feb 8th 2008 7:52PM]

CB17 = own3d