With all the talk of 4G recently, you’d think 3G was being put out to pasture. Not so, as is evidenced by AT&T’s upgrade to HSPA+ and the emerging M2M market, which 3G networks are likely to shoulder going forward. And there’s voice, an application that carriers have yet to deploy over their 4G networks, so for now at least, placing a call will be 3G’s job as well.
Secondary use cases aside, carriers still see their “old” data networks as a viable product for at least the next five, if not 10, years. If that’s the case, it looks at first glance as though the GSM carriers, namely T-Mobile USA and AT&T, might have some advantage over Sprint and Verizon Wireless, which haven’t yet upgraded their CDMA networks to Rev B. But is it really that black and white?
3G Still a "Great Product "
Verizon Wireless has said again and again that it does not plan on upgrading to EV-DO Rev B, emphasizing that its LTE coverage, which is deploying at 700 MHz, will be sufficient to meet its users’ needs.
Nicola Palmer, vice president of networks for Verizon Wireless, repeated her company’s position that it wouldn’t need an interim step to LTE. “Specifically, we didn’t want to take a half step, and we really looked at Rev B as that half step,” she said.
Admitting that rolling out LTE is “a little bit science, a little bit art,” Palmer says that Verizon is pursuing an aggressive deployment strategy, which will cover 70 percent of any market where it deploys right off the bat. “Performance on our Rev A network is good. Customers have been having great performance there, so we overlay the POPs, build out rapidly, and that negates any need to go to Rev B.”
Palmer says Verizon will continue to invest in 3G, as well as roll out more 3G devices. “Frankly, we’ve invested quite a lot of money in staying ahead of the data growth on 3G, and we anticipate doing the same next year,” she says.
Sprint agrees, echoing Verizon’s comments on the matter. Jay Bluhm, vice president of 3G networks development and engineering for Sprint, says Rev B was just unnecessary. "It's a much simpler and more beneficial for us to go directly from EV-DO Rev A to 4G, instead of an interim step that really doesn't give you the same data speeds and throughput," he says.
Bluhm says Sprint's CDMA network will continue to serve M2M applications, which don't necessarily need the kinds of speeds that 4G offers. But he also says that consumers will continue to use the network. "The 3G network will be around for quite some time as a fall back and also just for those customers that continue to stay on 3G-only devices and keep those devices for a very long time," Bluhm says.
HSPA+ and the 4G Label
So the prevailing opinion with the CDMA operators in the United States is that their 3G networks are just fine, with 4G offering enough off-load to keep 3G healthy. But how does that weigh when one considers the impressive HSPA+ speeds boasted about by the GSM operators?
Phil Solis, research director for mobile networks at ABI Research, says the idea that AT&T might have some advantage over Sprint and Verizon's 3G networks in the near-term because of its HSPA+ upgrade is not quite right. Because T-Mobile rolled out its 3G network so late, it was able to deploy base stations that can support 64 quadrature amplitude modulation (QAM), whereas AT&T's base stations are older and can only support 16 QAM.
The difference in how much QAM a base station can support means a big difference in what kinds of speeds the two networks will offer. While AT&T is currently claiming theoretical speeds of 14 Mbps, T-Mobile is seeing up to 21 Mbps and that disparity continues if the two GSM carriers pursue the evolution of HSPA+.
Solis says the 4G label just confuses things. “I think pretty much everyone in the industry is in agreement that WiMAX and LTE are the true 4G technologies,” Solis says, adding that the International Telecommunications Union’s efforts to define 4G may not have much bearing when it comes to how carriers like T-Mobile market their networks.
T-Mobile apparently feels confident enough in the evolution path of HSPA+ to call it 4G, suggesting it will hit 42 Mbps by next year and pegging peak rates at 168 Mbps at the far end of HSPA+'s evolution. AT&T is less likely to go there given its base stations and the fact that the carrier will increasingly want to keep focus on its impending LTE deployments.
“Up until now, we haven’t seen really any application breakthrough that says: This is something you can only do on 4G," Solis says. "I think T-Mobile is playing on that right now. You can pretty much do anything on their HSPA+ network that you can do on a 4G network today, thus they took the leap to say it’s 4G.”
Spectrum Makes the Difference
Michael Thelander, CEO of Signals Research, says Verizon and Sprint were probably justified in skipping Rev B. Still, he expects deployments of the technology from carriers like KDDI in Japan to drive economies of scale.
Thelander makes the point that Verizon’s 700 MHz deployment really does make a difference, whereas Sprint’s WiMAX, which uses Clearwire's network, and the high frequency at which it’s deployed, could mean Rev B is an option at some point.
Joe Lawrence, vice president of marketing for the CDMA Development Group (CDG), says the trick for those carriers rolling out LTE and WiMAX is going to be creating differentiation.
“The way it’s looking right now, with this demand for mobile broadband, the carriers need to grab as much spectrum as they can. And depending on what spectrum you can get your hands on, that’s going to determine what technology you deploy,” he says.
Even if operators don’t go with Rev B, there are ways that carriers can improve their CDMA networks going forward, Lawrence says. The CDG went to all of the CDMA carriers in North America and asked them what exactly it is that they need if they're not going to upgrade to Rev B. The carriers came back and said they wanted simple software that will allow them to add broadband capacity when and where they need it. The resulting solution is called DO Advanced, a software upgrade that allows the carriers to optimize their existing 3G networks.
"While you might not hear the big guys talk about it, inevitably they will implement these smart network techniques on their 3G broadband networks to be able to keep up with the demand," he says, adding that as the demand for LTE continues to grow, so too will the demand for 3G. "They'll have to invest in both networks to keep up with that demand."
An Ecosystem of Networks Emerges
At 4G World in Chicago this year, the consensus seemed to be that regardless of what kind of network the end-user is on, whether it be HSPA+, WiMAX, LTE, CDMA or Wi-Fi, they’ll be switching between various flavors depending on their location. In urban areas, they might get great coverage on WiMAX or LTE, but outside of those areas, fall back could be anything from 3G to Wi-Fi.
“We have this viewpoint that no one technology fits all,” Lawrence says, referencing KDDI as one example of a carrier that understands that the only way to meet the rising demand for data is to come at it from all angles.
“Basically, KDDI’s ploy is, I’m going to throw everything and the kitchen sink at this demand. I’m going to use Wi-Fi, WiMAX, 1X, EV-DO Rev B, LTE, wireline, broadcast channels… They’re going to use everything they can to meet the demand in Japan. I think you’re going to see multiple network types fulfilling the tremendous demand we’re seeing in broadband.”