Bux,
Re: CDMA Evolution to 3G Services (The Scenic Route of Migration)
<< The following article is a refreshing change from the typically politically motivated stories we have been exposed to. >>
This one is in the same vein as the excellent and informative article you posted, IMO. It adds some background and detail that I personally was unaware of (including comments on the importance of Motorola's X Plus and 1XTREME contributions).
The original article contains charts and tables not reproduced here.
>> "Taking CDMA’s Scenic Route To 3G Services"
By Mark Dziatkiewicz
November 20, 2000 Issue
Wireless Week
wirelessweek.com
Taking a trip often involves choosing between a direct highway route or the scenic back roads. Both ways get you there eventually, but the out-of-the-way path can open your eyes to sights and experiences you would otherwise miss.
So it is for CDMA operators on the road to third-generation services. Choice is the operative word for CDMA operators, who are looking for cost-effective, standardized alternatives to reaching their 3G wireless objectives.
The CDMA road to 3G has taken a few turns during the past several months. And while carriers could take the direct route on highway 3xRTT, the scenic route is getting more interesting every day. If speed isn’t the only priority, it could just be the best way to go.
Who Wants Dick Tracy?
Third-generation wireless is all about hard-to-believe data rates and outrageous applications like a Dick Tracy video on your wrist. But if 3G services can’t be offered cost effectively to the subscriber, does it matter what the technology is capable of?
In May 2000, the CDMA community took steps toward getting more bang from its cdma2000 upgrade bucks by adopting a high-speed data-only addition dubbed 1xEV-DO, or 1x Evolution-Data Only. 1xEV-DO offers most of what 3xRTT does but accomplishes it within the existing 1.25-megahertz CDMA carrier.
For operators, that is a big step toward cost-effective high-speed data because it avoids the hardware/software upgrade changes required with the 5-megahertz carrier of cdma2000 3x. "If the reason to go to 3xRTT is to provide high-speed data, but you can use the same bandwidth with 1xEV-DO and use less spectrum, it will give you a cost advantage," says Yusuke Higashi, CDMA product line manager at Nortel.
1xEV-DO evolved from the Qualcomm High Data Rate proposal proffered in late 1999. HDR demonstrated peak data rates above 1.8 megabits per second in a 1.25-megahertz carrier, but it was a data-only technology.
Motorola, on the other hand, announced its own competing proposal at CTIA 2000. Its technology "1X Plus" offered first phase data rates of 1.38 Mbps peak throughput on a single 1.25-megahertz CDMA carrier, later moving to 5 Mbps in a fixed environment.
In the end, the CDMA community chose the Qualcomm approach with modifications. Time to market and catching the high-speed mobile Internet wave were motivating factors to go with HDR, according to Ed Chao, Lucent’s senior manager, 3G and data product management. The opportunity to offer high-speed data rates that met or even exceeded those capable in GSM and TDMA 3G proposals -in a similar time frame and without costly hardware changes - was too good to pass up.
"Qualcomm had been working on [HDR] for over three years," Chao says. "They had been trialing the technology in San Diego and had done a lot of proof of concept." Because Motorola’s technology proposal was not as far along, the CDMA Development Group and its members opted for Qualcomm’s approach.
But 1xEV-DO isn’t an exact clone of HDR. Because HDR was designed for data only, it required a separate CDMA carrier with no voice integration. While the second carrier was something operators were willing to live with in the short term, the lack of voice integration wasn’t.
Relying on a separate carrier would have necessitated stand-alone devices on a high-speed data network, separate from the carriers’ 1xRTT voice networks. The business case wasn’t as appealing to operators, "because if they can’t inter-work and communicate there is less value for operators to provide the services," Higashi says.
For that reason, a number of modifications were made in the evolution to 1xEV-DO, including one to increase traffic throughput and another for interoperability with 1xRTT. The latter acknowledges the direction and importance of Motorola’s 1X Plus technology.
In an integrated solution, a terminal could offer both voice and data with hand-off capabilities. For example, the subscriber could be involved in a high-speed data session using the 1xEV-DO network when a voice call comes through. The data session then could be handed off to a 1xRTT CDMA carrier so the data session continues and the subscriber picks up the voice call, without involving two separate CDMA carriers.
"The implication is that the high-speed data application in progress might get flow control because 1X only supports a 144 kilobit-per-second data rate vs. the HDR rate," says Phil Hester, director of technical product marketing at Ericsson.
Providing the multiservice capability was an important concession and strategic modification. In some respects it offers the best of both the voice and data worlds. With 1xEV-DO, the separate spectrum allocation takes advantage of data’s characteristics to provide maximum efficiency, which is a nod to Qualcomm’s original premise. And the inter-working between voice and data reflects the operators’ wishes and market realities.
The phase one standard was scheduled for publication at the end of October, enabling field trials in January 2001. Manufacturers suggest commercial deployments will commence during 2002.
Focusing On Phase Two
At the same time, work continues on phase two of 1xEV, where the objective once again is to integrate phase one capabilities on the "same"CDMA carrier while retaining the ability to maintain packet data services on a separate carrier.
Peak data rates get bumped up to 1.25 Mbps in both directions (see chart) with average rates of 600 kbps in a fully loaded system with all users moving at vehicular speed. Phase one provided asymmetric data rates in all categories.
Other objectives in phase two include three modes of traffic delivery - real-time, non-real-time and a mixture of the two - and another doubling of voice service spectral efficiency. But most importantly, the standard should provide graceful evolution from existing CDMA platforms with minimal cost upgrades.
It’s a tall order, but since time to market isn’t as critical, the requirements have spawned three proposals to date, coming from Motorola, Lucent and LinkAir.
Motorola partnered with Nokia and repackaged 1X Plus as 1XTREME. "Integration of voice and data is the preferred method, so we’ve refocused our efforts on phase two requirements," says Neal Campbell, director of CDMA product operators for Motorola’s Network Solutions Sector.
One of the primary challenges in the integrated approach is packetized voice and handling the basic differences between voice and data. Voice needs a continuous stream while data uses bursts, so allocation issues exist.
Combining both types of traffic within the same bandwidth will require unique approaches in order to achieve cost efficiency while maintaining the necessary voice and data throughput. There seems to be agreement "that integrating voice and data is the right thing to do going forward, but no one is sure of the right way to do it," Nortel’s Higashi says.
Motorola could have an early lead with phase two because it’s been working for a while on an integrated approach. But still there are a number of requirements that 1XTREME would need to support, "and some of those things are pretty dicey," Ericsson’s Hester adds.
Ericsson doesn’t have a submission of its own, so it is closely examining what’s already on the table. Hester says one of the toughest requirements is to provide backwards compatibility with everything CDMA offers to date. And providing another doubling of voice capacity isn’t trivial either.
He indicates that Motorola uses some incredible modulation schemes to achieve its 5 Mbps burst capacity but suggests that throughput often suffers as a result. And there also are concerns about the required proximity between base station and terminal to support these extraordinary data rates.
Nortel isn’t bound to any proposal so far either but intends to see the process move forward in a way that is beneficial to everyone. Higashi points out that the current proposals deal with the radio portion only, and that alone won’t give you a cost advantage. The overall network solution is just as important.
Lucent submitted its proposal to the 3GPP2 (third-generation partnership program 2) standards group last month, according to Chao. Lucent was an early Qualcomm HDR supporter and is working closely with Motorola and Nokia to resolve differences between proposals.
The 1xEV-DV dark horse is LinkAir, a startup founded by one of the inventors of CDMA. LinkAir patented LAS-CDMA (large area synchronized-code division multiple access), which provides up to six times greater capacity than that of cdma2000, and the company hopes to position the technology as CDMA’s next evolution.
With Internet protocol technology at its core, LAS-CDMA already incorporates 1xEV-DV’s integrated voice and data requirements. And based on patented spreading technology, it eliminates the interference-limiting aspects of CDMA technology, according to a company white paper.
It also promises backward compatibility with existing wireless systems such as IS-95, IS-41 and GSM Mobility Application Part. But despite being a more efficient CDMA, some suggest it could have difficulty maintaining the backward and forward compatibility due to the nature of its implementation.
Whatever direction 1xEV-DO takes, it likely will undergo a few iterations before a final solution emerges. But an overall commitment exists on the part of manufacturers and operators to drive CDMA technology forward, while driving costs downward.
No single road exists to implementing 3G technology, particularly for CDMA operators. But sometimes the scenic route is simply the better way to go. <<
- Eric - |
