AT&T & LTE vs. T-Mobile & HSPA+ = Fight!

"He pulls a knife, you pull a gun. He sends one of yours to the hospital, you send one of his to the morgue."

I've always thought that famous line from "The Untouchables" went a long way towards summing up the U.S. wireless industry. No, I'm not comparing my 3G service provider to a mobster (though, I believe consumer groups have). Instead, replace the weapons and settings for wireless technologies and you get the picture. GPRS, CDMA2000 1X, EDGE, EV-DO, HSPA. The competitive back-and-forth that began long before low-priced smartphones drove mobile data traffic led to the continual deployment of increasingly efficient radio access network (RAN) technologies, each promising a better, faster user experience.

For the most part, these technology evolutions have been pre-ordained. GPRS begat EDGE which was followed by UMTS and HSPA. CdmaOne led to 1X, then EV-DO rev. 0 and EV-DO rev. A. Here on the eve of 2010, however, we come to a virtual fork in the road. Having deployed an HSPA network, operators could move today on HSPA+ upgrades (promising upwards of 20 Mbps in a 5 MHz channel) or wait for LTE (promising similar speeds initially) to materialize over the next few years. In many parts of the world, the two are seen as far from mutually exclusive: operators worldwide are moving on HSPA+ upgrade while plotting a course towards LTE. Here in the U.S., however, we have a different situation with T-Mobile USA moving aggressively on HSPA+ and AT&T Mobility prioritizing LTE.

Now, I'm just a network guy. I can walk you through the myriad HSPA+ permutations (with MIMO, with 64QAM, with both, with both and multiple carriers) as well as the implications of a chatty X2 interface on an LTE networks Evolved Packet Core. Operator decisions on network technologies and deployment timing, on the other hand, are beyond my limited comprehension. Since that's never stopped me in the past, however, I figure it's worth weighing in on the two strategies--if only because they're so divergent.

If you're about to skip to the end of the column to see my proclamation of a winner, let me save you some time: I like the T-Mobile strategy better. Sure, the company admittedly moved too slowly on 3G upgrades. Rolling out HSPA+ as quickly as transport assets will allow, however, should deliver the data services and devices (including support for today's HSPA devices) end-users are looking for in a shorter time frame than waiting for LTE to mature.

Like any analyst argument, however, there are some caveats.

On a very basic level, neither operator really has a choice. T-Mobile, for example, might prefer to jump straight to LTE. Here in the U.S., however, it doesn't look to have the spectrum resources to keep its 2G and 3G networks running while simultaneously deploying a new network. AT&T does have the spectrum. It also has the RAN assets that should be able to support an upgrade to HSPA+. Yet, to the extent that it's perpetually locked in a battle for mindshare and subscribers with Verizon Wireless, it must broadly match or surpass the big V's plans--and that means deploying LTE.

Beyond the data rates or devices offered in the near-term, simply offering 4G or LTE services will carry marketing value. It's unclear how much value Sprint Nextel is getting from the use of the term for its WiMAX network. Regardless, moving (relatively) early on LTE should help AT&T to mitigate the 3G coverage issues that have plagued it with HSPA while giving it a better claim to the mobile broadband speeds Verizon Wireless will doubtless start promising next year. And the year or so delay over Verizon's LTE launch? That should give the market time enough to work some bugs out of the technology and develop devices beyond USB dongles.

Ultimately, end-users get to decide which operators and mobile broadband services make the most sense for them--and they don't care about technologies. Your average 3G user probably can't tell you if they're on an HSPA, EV-DO or TD-SCDMA network. Why? Again, they don't care. What they do care about in a mobile broadband world is the quality of their services (throughput and latency) weighed against the cost of those services and the devices available to access those services. Services! (I felt it was worth repeating again to make my point)

So, why do I side with T-Mobile in this particularly battle? In the near-to-medium term, a strategy based on HSPA+ should deliver sexier data rate promises and a broader set of cool devices. It also plays into T-Mobile USA's longstanding strategy of waiting move on a technology until it can leverage an established (cost-reduced) ecosystem. To be clear, I'm under no illusion that a 21 Mbps HSPA+ upgrade will do much to significantly boost cell capacity and few phone-based applications will benefit from 10+ Mbps peak rates. I also know that T-Mobile will want to move on LTE someday when it gets more spectrum or can move its 2G users to 3G. From a device, marketing and overall end-user experience standpoint, however, I see the close-in future belonging to HSPA+...luckily for AT&T, if LTE development stalls or takes longer than expected, it can flip-flop again and re-prioritize HSPA+.

Peter Jarich is an analyst with Current Analysis [1].