When a person believes something for which there is no evidence, we call that blind faith. But when someone refuses to believe what has already been demonstrated - like the competitive advantages of cdmaOne™ - that's called wearing blinders.
After years of rancorous theoretical debate, 1997 was the year CDMA was put into widespread practice. Stunned critics watched silently as the technology they once described as an attempt to "violate the laws of physics" went live in city after city, amassing millions of subscribers. IS-95 technology, now known as cdmaOne, stepped into the bright sunshine of the free market.
Certainly cdmaOne had its birthing pains. And there are still several areas in need of improvement. So when Telephony magazine set out to determine whether CDMA was living up to its promises, they had the right idea. Unfortunately, reporter Steven Titch got swept up in the very same "holy war" he professed to abhor. Instead of an objective review of cdmaOne's achievements and shortfalls, he delivered a long-winded sermon teeming with vague allegations and thinly-veiled ad hominem attacks.
Shortly after the article appeared, the author assured me his only motive had been to provide readers with a more balanced perspective. That's certainly a noble cause. But it hardly squares with the article's "Blind Faith" title, or the "Hype" caption directly under a photograph of Qualcomm CEO Irwin Jacobs, or the constant insinuation that Qualcomm lied to customers-particularly since the article failed to identify a single CDMA operator willing to corroborate the charge.
"Blind Faith" begins with what has become the most familiar allegation against CDMA: that it does not deliver the capacity promised by Qualcomm. Qualcomm stated CDMA would provide 10 to 20 times the capacity of the North American analog standard, AMPS. Yet several U.S. operators report cdmaOne delivers about 6 times AMPS' capacity.
Is there a discrepancy? Qualcomm promised 10 to 20 times AMPS' capacity just when it looked like operators were going to standardize on 8 kilobit per second (kbps) voice encoders (vocoders). But the U.S. market firmly rejected the audio quality of a competing 8 kbps digital technology so-realizing they had a big capacity advantage-U.S. CDMA operators wisely upgraded to the 13 kbps vocoder. The 6 times AMPS' capacity reported by carriers employing 13 kbps vocoders is roughly equal to 9 times AMPS at 8 kbps. In addition, Asian cdmaOne operators who stuck with 8 kbps vocoders have publicly reported achieving 9 to 10 times AMPS' capacity. While these measurements barely reach the low end of Qualcomm's promised range, we must not forget they come from the very first commercial cdmaOne networks. Not a bad start.
Many of the article's allegations were based on competitors' opinions, statements taken out of context, or the author's far-fetched interpretations. A good example is the assertion that Lucent has been "quietly breaking ranks" with Qualcomm, telling customers not to expect more than 6 times AMPS' capacity. Again, this is at 13 kbps. And we must also remember Lucent and Qualcomm are competing infrastructure manufacturers. But most devastating is the fact Lucent has repeatedly told customers they can expect IS-95 CDMA systems to deliver twice as much capacity in fixed applications. Based on the original 8 kbps yardstick, that's roughly 18 times AMPS' capacity-close to the top end of what Qualcomm promised.
Next, the author charged that IS-95 networks "can be difficult to optimize amid rapid subscriber growth." It is hard to argue with that statement. It's difficult to optimize any cellular radio network amid rapid subscriber growth-not that cdmaOne operators are complaining, mind you. But according to the German GSM operator T-Mobil, who recently trialed a Motorola CDMA system, IS-95 works reasonably well using the manufacturer's default settings. Even so, it's true cdmaOne operators are just learning how to optimize their networks, and we can expect them to become more proficient as they acquire experience.
This was followed by the claim that CDMA cannot be overlaid on AMPS networks "in a 1-to-2, let alone 1-to-5, [base station] ratio with any expectation of quality coverage." Yes, there is disagreement among cdmaOne carriers about the best way to overlay an AMPS network. Some lean towards a 1-to-1 ratio of CDMA-to-AMPS base stations in urban centers. One operator suggests a 1-to-1 ratio of CDMA-to-AMPS base stations-but only in suburban and rural areas. Most operators agree, however, that it takes significantly fewer CDMA base stations to overlay an entire AMPS network.
The opening attack concluded with the observation that IS-95 CDMA "is far from being as mature as other digital wireless technologies." That's a reasonable statement, except the reporter implied cdmaOne vendors were claiming otherwise-again, with no evidence. The cdmaOne industry can be proud of how quickly it has brought a relatively ambitious technology to market, and more important, how quickly it has established cdmaOne as a legitimate competitor to technologies that enjoyed at least a three-year headstart.
Read further, and you'll find allegations in "Blind Faith" bordering on the absurd. For example, the author suggests there is a discrepancy between what Qualcomm promised and what CDMA delivers that "makes carrier business plans incorrect by several orders of magnitude." We are apparently to believe that carriers are continuing to deploy cdmaOne in city after city despite discovering it costs many times more than anticipated. Unfortunately, there is no way to know precisely what the author meant by "several orders of magnitude," because he presented no cost data.
Then we come to what was for me the moment of truth. "Blind Faith" informs us that coverage concerns delayed the commercial introduction of CDMA. Someone better tell Sprint PCS, who rolled out service in 134 cities over a twelve month period, and PrimeCo, who pulled off the largest simultaneous launch in the history of wireless. Or better yet, tell AT&T Wireless Services, who managed to roll out service in a paltry 10 cities during the same period using a competing technology. The empirical evidence that cdmaOne has a coverage advantage is so overwhelming, you have to be wearing blinders not to see it.
"Blind Faith" repeatedly misleads readers about other issues. For example, CDMA cell breathing is presented as an unexpected problem. Actually, cell breathing has been discussed for years and is one of CDMA's advantages. It hastens the launch of commercial services, and acts as an automatic load balancing mechanism as traffic grows. "Blind Faith" also asserts that cdmaOne systems can't do hard handoffs. This is simply false. And in a sidebar entitled "Count the cell sites," the author uses raw numbers from Tampa, Florida to suggest CDMA requires more cells. But he provides no data on the actual coverage and capacity of each network.
In the section titled "A Jihad is Launched," the author finally tells us what's really bothering him. He opines that IS-95 CDMA is mainly a nationalist reaction to the success of Europe's GSM, and accuses the CDMA industry of attempting to "block encroachment of GSM... into the U.S." But the exact opposite is true: Europe mandated GSM to the exclusion of IS-95 CDMA, while GSM networks are operating in dozens of cities across the U.S.
In researching "Blind Faith," the author placed all of his faith in people who have been crusading against IS-95 CDMA for years. One is consultant Richard Russell, who presented a paper at a GSM conference in April, 1996 in which he compared the CDMA industry to a British woman who claimed to have given birth to seventeen rabbits. In retrospect, he should have compared her to himself. In the very same paper he predicted the commercial launch of CDMA as an AMPS overlay would be delayed up to 18 months. But AirTouch launched an AMPS overlay service (Powerband in Los Angeles) just a few weeks later. Mr. Russell-like that British woman-was off by seventeen.
Sorry Steve Titch, but your sources were tainted, and your claims of objectivity proved hollow. The free market is the ultimate truth test: cdmaOne is off to the fastest start of any cellular technology in history. You couldn't name one manufacturer or operator who is unhappy about choosing cdmaOne. In fact, the only people complaining about cdmaOne these days are nervous competitors. And, oh yes-those who don blinders for religious reasons.
About the Author
Ira Brodsky is President of Chesterfield, Missouri-based consulting firm Datacomm Research Co. His new book, Wireless Computing: A Manager's Guide To Wireless Networking, is available from Van Nostrand Reinhold (New York City, NY) by calling 1-800-842-3636. He can be reached via e-mail at firstname.lastname@example.org.