3G Networks

A Lesson in How NOT to Do Things

Francis Vale


The trick to making a successful embedded everywhere system is that it should hardly be worth thinking about. Like zippers, for example. We don’t think twice about them, except when our clothes get stuck in them at the worst possible time in the worst possible place. But this mostly ignored piece of ingenuity took twenty years to develop and another twenty years to achieve widespread market acceptance. Maybe it’s their fear of the prolonged zipper effect that‘s subconsciously haunting mobile Telcom execs, prompting them to rent $40,000-per-week yachts for promoting their new 3G wireless wares during the February 2001 3GSM World Congress in Cannes, France. And was it market motion sickness that prompted the president of NTT DoCoMo of Japan to say during the Conference that by 2010, ten million pets would be suited up with 3G data gadgetry, not to mention those 60 million 3G devices embedded in bicycles he also predicted? However, Hitachi indicated that if this new 3G boat had in fact set sail out of the Cote d’Azur, it wasn’t on it. Ian Smythe, manager of the company’s mobile components group, groused "I can’t say that we have received any orders for 3G technology yet."

And while NTT partied onboard, Motorola, one of the world’s largest mobile Telcom suppliers, was sighted slipping beneath the profitability waves as it said it would be posting a quarterly loss, its first in fifteen years. And then Olav Ostin, UK managing director of the global venture capital firm ETF Group Half, threw cold water over everybody when he said fifty per cent of the companies exhibiting at this chi-chi World Congress would go bust in the next 12 months. He then added that those most at risk were the wireless content providers. (How do you spell WAP?) Finally, Europe, which is supposedly a generation or two ahead of the U.S. in deploying advanced mobile wireless systems, is still struggling to get its "2.5G" networks on the air. Increasingly, it looks like 3G is becoming a slow boat to China instead of being an overnight revolution in high speed wireless data communications and mobile embedded devices.

While dot com fever raged and pervasive Internet access was seen as the road to never ending IPO riches, high speed mobile wireless data networks, like 3G technology, were seen as a sure thing, yet another massive market piston propelling the Internet economy fast forward. But then a NASDAQ run by a crazed Captain Nemo torpedoed the economic engine room. Nonetheless, it apparently hasn’t stopped Telcom execs from frantically rearranging the deck chairs on this titanic devourer of investment capital. Sir Richard Branson, of Virgin fame, effectively said damn the 3G icebergs and full speed ahead when he announced that his new mobile company would soon be investing $1 billion building Virgin Mobile U.S. But large as this number sounds, it’s a pittance compared to the $180 billion it’s projected to cost for global mobile operators to go 3G. And even the rosiest of industry projections, such as by Orange, Europe's number two mobile operator, indicate that 3G profitability is far off into the future. Orange says the ink will run red for at least four to seven years after its own service launch in 2002. But that may slip to 2003. DoCoMO was originally scheduled to begin its 3G long march in May 2001, but muffed its brazen marketing dare. If it had met its launch target, DoCoMO would have become the world’s first provider of third generation high-speed wireless data service.

When put into historical perspective, we see that DoCoMo’s seemingly naïve 3G actions and putative pussy and pooch predictions make perfect sense, as well as gain a better understanding of why the wildly successful DoCoMO has been such an "overnight" success. It has been little remarked that NTT/DoCoMo's i-mode always-on handheld Internet terminals are based on TRON-specification technologies--ITRON and JTRON. From a cold start on February 22, 1999 when DoCoMo first began offering i-mode Internet access via cellular technology, its TRON-toting i-mode customers numbered more than 10 million come August 2000, beating even Goldman Sachs’ rosiest predictions by more than six months. Contrary to all conventional wireless Net revenue wisdom, NTT DoCoMo is currently raking in billions of TRON-enabled, un-WAPped, highly profitable yen per year. "We set up the Internet business model for i-mode, rather than set up a wireless Internet model," says Takeshi Natsuno, media director of gateway business department. "We don’t need wireless specialist content providers." i-mode is implemented in all-HTML as WAP was not available at the time I-mode was being planned, which may, in retrospect, have been a good thing.

The origins of TRON date back to 1982, the same year as the eponymous Disney movie. It was then that Dr. Ken Sakamura, an unknown assistant professor from the Department of Information Science at the University of Tokyo, put forward a radical notion in a small subcommittee on microprocessors and operating systems at the Japan Electronic Industry Development Association (JEIDA). His bold recommendations led to the launching of the TRON Project in 1984. Its goal was, and still is, "total computer architecture" in which all the devices in the human environment are computerized and linked together. The 2001 DoCoMo 3G service rollout is thus the culmination/continuation of twenty years of infrastructure planning in Japan.

TRON-enabled devices were called "Intelligent Objects," and the BTRON-specification personal computer still remains the only one with a hypertext filing system. The BTRON, or "Business" TRON specification computer architecture, was also designed for platforms with limited resources and incorporated a new unified model for user operation, data management, and program execution. This new system concept was called the "real object/virtual object model," and it presaged the hypermedia functions of the yet to be invented World Wide Web. In addition, a subset of the BTRON3 specification, which is called "micro-BTRON," was recently defined for use in PDAs. BTRON was unveiled in 1988, a seemingly distant time when the Internet had yet to come into existence, which makes for a delicious irony. The only ingredient lacking to make Japan’s TRON "real" was the connective glue supplied by the USA-backed Internet and TCP/IP protocols. Regardless, every time you hold a Silicon Valley hyped PDA in your hand or hear about some new object-oriented this or that mobile system, they are likely twenty year old ideas whose ‘innovators" have no notion they are probably reinventing a made-in-Japan wheel.

By market cap, even though its shares have been beaten down 25% from 2000 prices (as of this writing, Spring 2001), NTT DoCoMo is now Japan's biggest company and the world's biggest mobile-phone company. Unlike the West and its business plan du jour mentality (and the never-ending peccadilloes of AT&T, which has been angling to buy 20% of DoCoMo for about $10 billion) NTT/DoCoMo has simply been following a plan, one that was meticulously laid down and continually refined over many years. Most analysts in the West fail to grasp this essential fact, although they are correct when they worry about DoCoMo being able to successfully export its business model elsewhere in the world. In many ways, DoCoMo and TRON are a uniquely Japanese construct.

Twenty years seems to be a magic number, if zippers and TRON-gestation are any indication, so maybe mobile 3G networks and pervasive, always on, Net-device attachment are finally upon us. The question is, will it take another two decades to zipper up this new business? The answer largely depends on who and what ends up winning the 3G technology and business wars, an outcome that becomes more uncertain with each passing day.

Second generation GSM (Ground System for Mobile Communications) systems use TDMA (Time Division Multiple Access, selected in 1989 as a digital cellular standard) with dedicated channels for each side of a voice conversation and a network optimized for voice traffic. Accordingly, 2G GSM networks bear a strong resemblance to the wire-line telephone infrastructure. Data users are mostly treated as wasteful, second class citizens in the 2G world, lucky to get even a measly 9.6 kb/second per time-slot. Moreover, since a 2G channel is dedicated as long as a call is maintained, regardless if users are talking, GSM circuit-switched channels are only 40% efficient. But in a new Internet dominated world with its data available everywhere demands, this simply will not do. Hence, the new race to 3G high-speed data/efficient voice networks, with apparently, a very prolonged pit stop at 2.5G.

When the marketing dust settles, the so-called "2.5G" networks are primarily differentiated from older 2G setups by higher data throughput. Anything that hits a data rate above 56 kbps apparently earns the moniker 2.5G, although always on service is often touted as being part of the 2.5G market fabric. 2.5G networks use six conflicting and incompatible systems to attain these higher numbers: HSCSD, GPRS, EDGE, EGPRS, IS-136B/HS, and cdma2000-1x. As of late, no less than four of these; HSCSD, GPRS, EDGE, and cdma2000-1x; have been getting most of the TELCO attention.

HSCSD allows concatenation of multiple time-slots to increase data throughput, but it still retains the original dedicated voice channel architecture of GSM. Basically, HSCSD is a slight modification to the original GSM coding scheme, coupled with multi-slot (channel) operation. Combining these two techniques permit user data rates to double to 14.4 kb/second per slot, or even shoot up to 57.6 kb/second if four dedicated time slots are combined. Thus, a three or four-time slot HSCSD network is a 2.5G rig, even though it’s not a packet network overlay system.

However, time slots are limited and thus incredibly valuable for the bread and butter business of cellular operators — voice. Moreover, the use of a dedicated channel, although a natural extension of wired telephone networks, is a poor choice for the short, bursty nature of a data network. Hence, HSCSD service is typically available only where cells are not fully loaded or during off-peak hours. To counter these thorny issues along came GPRS (General Packet Radio Service), another member of the 2.5G family.

GPRS creates gets rid of the dedicated slot scheme of HSCSD and substitutes shared channels capable of transmitting data packets from multiple users on each channel. GPRS is a separate packet network overlaid on the original circuit-switched voice network. A shared GPRS channel is rate adaptable, capable of adjusting itself to varying signal propagation and interference conditions. As a consequence, four possible data rates are available per channel, ranging from 9 kbps to 21.4 kbps. And like HSCDS, multi-slot operation is also permitted, providing, theoretically at least, data rates as fast as 171.2 kbps by using eight time slots under the best possible conditions. In truth, that 171.1 kbps number is only available under ideal conditions — you live in an aerie next to a cellular tower and you are the only person in the world on the network

Live GPRS simulations done by the industry indicated maximum user data rates under realistic conditions to be only about 40 kbps. Worse, in heavily loaded cells, GPRS delivered only about a 2x improvement in data rates compared to plain old GSM, primarily due to the use of packet-switched channels. This is way, way off the 170 kbps high water mark. Naturally, none of this insider info was brought to the public’s attention by the Telcom industry.

And so, when British Telcom’s Cellnet unit launched the world's first commercially available GPRS service, surprise, surprise, it offered a miserable throughput of just 26 kbps. Faced with the harsh light of day, GPRS stalwart Nokia now had to admit that realistically, maximum data rates of about 43 kbps were going to the norm. Ericsson was a tiny bit more upbeat and thought that 56 kbps would be achievable using GPRS. Either vendor’s revised estimates should still be met with some skepticism, however. The Cellnet performance was far removed from the 115 kbps to 170 kbps numbers that GPRS backers breathlessly promised users, even though everyone close to the situation knew that quoting such high rates was totally off the wall. In GPRS practice, cells quickly become overloaded and no way can a single data user expect to be allocated all eight time-slots. Truth in marketing has historically seemed an oxymoron in the Telcom vocabulary.

Interestingly, 43kbps also approximates the throughput achievable from HSCSD when using three channels, and at least one operator, UK-based Orange, plans to offer both HSCSD and GPRS as new 2.5G services. Orange currently offers HSCSD - although only at 28.8kbit/s - and intends to roll out GPRS later this year.

It’s easy to see the business logic behind Orange’s thinking. Unlike the simple base station modifications required by HSCDS, GPRS requires both hardware and software modifications that delve much deeper into the infrastructure. Estimates for converting an existing GSM system to GPRS are in the range of $8,000 per base station, so, for example, one UK operator has estimated the conversion costs to be $50M for their 5 million subscribers. Still, $10 a head is not all that bad, presuming you can get them all hooked on various types of paid data services.

Significantly, GPRS, unlike HSCDS, enables always-on service, a feature in a packet-switched network that NTT's highly successful i-mode has shown to be more important to users than raw data rate. GPRS delivers faster connection-set-up time than GSM, permitting data niblets to be continually downloaded to a user's terminal. A form of GPRS, in fact, will be at the packet core of the base stations providing the always-on capability for 3G networks. The always on feature may be the deciding GPRS blow to HSCDS. DoCoMo’s Natsuno has established four criteria for wireless data success, be it 2G i-mode, 2.5G, 3G, or any other type of user service elsewhere in the world:

  1. It must be timely and continual, not just updated once a day, hence the power and attraction of always on service.

  2. It should be deep and permit easy drilling down to other levels.

  3. It should encourage repeat visits (e.g., games, stock quotes, sports scores);

  4. The user should be able to clearly see the benefit of using the service, (as opposed to being forced to wade through the miasma of vendor service hype).

As the packet-switched nature of GPRS allows billing either by minutes of use (for voice) or by volume of data transferred, service providers should be beating their marketing brains out to do, and not merely say, what Natsuno preaches. If they don’t, expect more articles wondering why DoCoMo is such a success and no else is.

With GPRS now proven to be only slightly better than HSCDS, a path was cleared for a third combatant to step into the 2.5G ring, EDGE (Enhanced Data Rates for GSM Evolution). EDGE is a higher data rate, logical extension of GPRS whose claim to fame is its supposed ability to deliver maximum user data rates in excess of 384 kbps. This feat is accomplished by changing the modulation format to pack more bits of information into each slice of frequency spectrum; theoretically retaining full system capacity while increasing achievable data rates. If by now you are somewhat wary about these much higher numbers, then good, you are learning fast, because the fly in the EDGE ointment is error rate correction, or maintaining acceptable quality of service (QoS).

As with the other competing systems, to get the fastest possible EDGE connection you have to live on top of the cell tower and be the only one on the system. The higher modulation levels of EDGE (a TDMA system) require increased signal strength at the receiver than GPRS. Move away from the base of the cell tower, and EDGE’s link quality control protocol will reduce the data rate to maintain a continually acceptable QoS, with co-channel interference being the primary rate-killing culprit. Long range, overcrowded cells, and/or lots of co-channel interference will quickly put paid to those 384 kbps you are paying a premium for.

However, EDGE’s link-quality control is significantly better than that offered by GPRS, as indicated by Ericsson simulations. According to Ericsson, even when measured at a comparable GPRS signal strength, EDGE delivers twice the data rate performance per slot. But as EDGE uses higher signal strengths than GPRS, data rates will keep climbing after the weaker GPRS signal poops out. Using the real world BT Cellnet GPRS experience as a guide, this probably means an average single slot EDGE data user will see about 45-60 kbps performance, and maybe a little higher in some networks. In comparison to GPRS, EDGE does seem to enjoy an edge. However, even with this improvement, performance is just about comparable to analog dial-up, so the broadband beef is still out there somewhere roaming the high frequency range. In GSM-based systems, an upgrade to EDGE will involve mainly software upgrades and estimates for upgrading to the system are in the range of $12,000 per base station controller, which are still reasonable.

After 2.5G comes 3G, the supposed last step in mobile communications, at least for traditional cellular voice operators. 3G is billed as being a broadband service, featuring packet-based transmission of text, digitized voice, video, and multimedia. Unlike 2G and 2.5G macrocell networks, 3G deployment locations will also include the deployment of microcells and picocells. And where an operator is not constrained by supporting an existing 2G network, a second stage 3G IP network architecture is truly all-IP.  This all-IP capability is the primary difference between a completely new 3G network and one that is an upgrade to a legacy 2G network.  Improved access to Internet-based applications through the use of IP-based mobile terminals, offering seamless end-to-end services, is now possible. For this type of new all-IP network, some sort of VoIP or voice-over-packet standard, such as H.323, will be used for voice services. If they become available, true all-IP 3G systems will have huge infrastructure and management planning implications for major end user organizations.

3G is supposedly an all-encompassing, worldwide single standard. But it’s abundantly clear that geopolitical/market/technology forces will prevent a single worldwide 3G standard from coming into being for some time to come. The 3G version of GSM is technically called UMTS 2000 (Universal Mobile Telecommunications System), but also goes by the moniker IMT-2000. W-CDMA (Wideband-CDMA) was decided upon by existing GSM equipment suppliers, who are mostly European and Japanese manufacturers, as the basis for 3G/UMTS. But in the standards process, another major contender to W-CDMA emerged, called cdma2000.

cdma2000 is backward compatible to the IS-95 standard, also known as cdmaOne. IS-95 is widely deployed in North America and in forty other countries around the globe as a 2G system. Often called just "CDMA (Code Division Multiple Access)," IS-95 is a global archrival to GSM-TDMA, and the two digital cellular systems, as world travelers know all too well, are incompatible. But W-CDMA went even further--It is not backward compatible to ANY 2G system, be it GSM or CDMA. The most discussed and debated parameter between the two dueling 3G systems is the system chip rate. W-CDMA uses a chip rate value of 4.096 Mbps, while cdma200 uses 3.6864 Mbps.  W-CDMA supporters claim as much as a 10% capacity improvement over that of cdma2000. Another key distinction is that CDMA2000 is totally CDMA, while W-CDMA uses GSM for voice, but CDMA for data.

The patent holder of CDMA is San Diego-based Qualcomm, who obviously has a very big stake in all this, as it expects to get royalties, estimated in the many billions, no matter which CDMA-based system wins the day. Qualcomm is periodically accused of playing hardball over the 3G intellectual property issue. The company claims to hold patents to all of the essential key elements to any given 3G CDMA standard. However, DoCoMo, Ericsson, Matsushita, Oki, Siemens and Interdigital have all contributed to the new W-CDMA standard, as have a host of lesser known companies. But Qualcomm adamantly insists, and most analysts agree, that its IP portfolio is wide and deep enough to ensure that it will get its royalties no matter which competing 3G system wins the day. (Regardless, there are likely to be some attempted legal ambushes on the way to the National Bank of San Diego.) So why, then, has Qualcomm been so vocally adamant that cdma2000 is superior to W-CDMA?

Because nothing in 3G W-CDMA is backward compatible with a 2G GSM- TDMA system, transition costs are in the billions. NTT DoCoMo’s new 3G service is reported to have cost $10 billion dollars to implement. This is not chopped liver. Moreover, 2G/2.5G operators also have to spring for a new 3G license. Five British 3G licenses alone cost about $33 billion dollars. And in Germany, six operators probably had to sell their BMWs to come up with the $46 billion in fees they collectively shelled out. The French, not to be outdone, are going straight to the bottom line guillotine, demanding a head chopping $4.6 billion dollars per license. And being French, simply being rich is not enough. The French 3G licensor also has to pass "beauty contest" where it’s the quality of the operator that counts, not the quantity of cash on hand. Recently, the 3G tulip mania seems, temporarily anyway, to have subsided. For example, license bidders in France warned the government to either cut the exorbitant fees in half or the country will face losing its G-string entirely. All this financial bloodletting has consistently pushed back 3G W-CDMA deployment until 2003 and maybe even into 2005. None of this made for a sunny day on the Cote d’Azur. After Qualcomm’s Chief Executive Irwin Jacobs noted at the February 2001 3GSM World Congress that W-CDMA would not be commercially viable for several years, Qualcomm shares fell to their lowest levels since October 1999.

So where’s the good news? Well, those "lucky" auction winners can take solace in the fact that the 3G standard also allocates spectrum in existing TDMA 2G systems, preserving and incorporating their legacy infrastructure investments. But that also means two things: 1) Legacy operators will not feel compelled to build new all-IP networks and 2) 3G handsets will have to support multiple formats, offering at least 3G, GPRS, EDGE and maybe HSCSD. This support will be a major impediment for manufacturers trying to design battery efficient, small, low cost handsets. In contrast, existing handsets will be interoperable with a 3G cdma2000 network, as it is backward compatible with 2G IS-95 systems.

Power users in Europe and Japan will be the first to spring for these big, costly, battery-sucking, 3G sets, as per cell data rates will range up to 2 mbps in stationary or pedestrian environments and up to 384 kbps for mobile users. However, like 2.5G systems, data throughput will decline with increasing numbers of users per cell. And as with 2.5G, 3G operators cannot guarantee a minimum set data rate to a user. In sum, those looking for QoS-reliant, mobile broadband services, like the highly touted 3G video services, will most probably have to look elsewhere.

And the 3G picture will be fuzzy for some time to come, because until complete 3G deployment takes place, interim solutions, based on TDMA and/or hybrid techniques, will also be evolving into 3G. Most of these interim solutions will probably be incorporated into EDGE. AT&T, for example, has already announced its intention to use EDGE for its new 3G networks, proclaiming its maximum data rate of 384kbps as 3G proof positive. But achieving that 384 kbps rate in the real world is suspect. Perhaps most questionable of all, the great 3G hope almost totally rests on wide spread WAP deployment and the Internet. Given the current disastrous dot com business climate, this is not exactly the kind of planning premise to bank billions on.

And if you are X-Files paranoid, you might also be asking yourself if the Europeans are trying to lock out U.S. manufacturers from the lucrative European wireless equipment market, and are also seeking ways to shave those huge Qualcomm royalties. The Europeans pulled a similar stunt once before, when they initially locked out "foreign" makers of 2G GSM-TDMA gear. Ironically, it was CDMA-based PCS in the U.S. that started a whole new ball game where everyone could play, and in the biggest market to boot.

In light of all this, is it any wonder that Mr. Olav Ostin said fifty percent of the companies exhibiting at the 2001 3GSM World Congress would be going bust in the next 12 months? Frankly speaking, it’s a huge mess. But then again, the market is also huge. Research firm Yankee Group predicts that there will be 60 million wireless data devices in America alone by 2005. With predictions like this flooding the marketplace, it’s no wonder that the big operators are so determined to lighten their wallets a billion plus at a time.

The long road to 3G data networks in the US was very different from that traveled by the highly unified European GSM operators. In the U.S., 2G D-AMPS (IS-136), typically called just TDMA, is the major digital service competition to cdmaOne (IS-95, or CDMA). But in terms of global 2G users, cdmaOne dwarfs D-AMPS, although AMPS proponents often insist that both analog 1G AMPS and 2G D-AMPS users should be counted together because of their infrastructure similarities. But even with industry inflated user numbers, it’s unlikely D-AMPS and its version of GPRS/EDGE (IS-136B/HS) will ever make it to widespread 2.5G deployment. It especially won’t make it to 3G networks because, like 2G GSM, it’s TDMA-only.

That leaves AMPS users looking for data access with Cellular Digital Packet Data (CDPD) service, which enables both 1G and 2G AMPS networks to carry packetized data alongside voice. It is deployed as an overlay to both analog and digital cellular AMPS networks without requiring any additional radio spectrum or degrading the capacity and quality of the existing cellular phone service. Only minor changes are needed to the radio base stations. CDPD uses either idle voice channels or dedicated data channels depending on network configuration. The raw throughput of CDPD is 19,200 bps but actual throughput after protocol overhead is only 10,000 to 12,000 bps, if that.

On the other hand, 2.5G CDMA operators in North America are going straight to cdma2000, claiming that IS95B, another 2.5G spec, can only go to 64 kbps. (This is not true. In August 1999 South Korean cellular operator SK Telecom used IS95B to launch the world's fastest cellular data service, enabling users to transmit data at up to 115kbps.) At least count, cdma2000 comes in at least three data rate flavors. Already in commercial service in Korea, cdma2000-1x offers a data rate boost up to 144-153 kbps, but is still way below 384 kbps EDGE systems. This variant can be deployed in existing spectrum. cdma2000-3X combines three CDMA 1.25-MHz radio channels, offering 384 Kbps outdoors and 2 Mbps indoors, for higher-performance services, and will probably require additional spectrum purchases. However, Qualcomm has offered a third, new option: HDR (High Data Rate), which is now referred to as 1XEV, signifying an evolution of 1X technology. 1XEV uses a 1.25-MHz CDMA radio channel dedicated to and optimized for packet data, with throughputs of more than 2 Mbps being claimed.

The architecture of CDMA2000 is fundamentally different from that of GPRS/EDGE, which use specialized, cellular-specific protocols. CDMA2000 instead uses PPP to link users to a PDSN (packet-data serving node) and Mobile IP to support customers roaming between CDMA2000 networks. If it lives up to the hype, the Mobile IP option will allow customers with unique IP addresses to actively roam in CDMA2000 networks.

Despite all the 3G compatibility advantages enjoyed by American cdmaOne operators, which should give them a significant edge over their overseas GSM rivals, the US seems determined to be the last major country to turn the 3G lights on. The United States doesn't plan to have 3G spectrum available for carriers until at least September 2002. At that tortoise rate, it may not be until 2004 or 2005 before that first U.S. 3G cell phone starts its annoying beeping. It was only in January 2001 that the FCC finally got around to asking for comment on the best frequencies to be designated for 3G services. This woefully late action was part of an initiative started by President Bill Clinton in 2000 to spur U.S. deployment of 3G services.

Some observers think this 3G tardiness is a good thing, citing, for example, that the US now has an opportunity to bypass 2G/2.5G legacy systems and go straight into "true" 3G systems offering all-IP services. But that ain’t gonna happen, Bunky. We have seen that AT&T, one of the biggest operators, is already legacy backpedaling, calling what’s basically a 2.5G system (EDGE) a 3G network. Moreover, Verizon, et al, are not going to sit idly by and let AT&T steal all the 3G thunder, even though it’s really just a 2.5G whimper. They will quickly follow AT&T’s lead and legacy 2.5G networks will sprout like co-mingled mad mushrooms across North America. Indeed, both Verizon and Sprint PCS have already said they plan to start offering 153 kbps cdma2000-1X service in 2001.

But there are other contenders for an all-IP xG network. One that drops W-CDMA and cdma2000 altogether is from Flarion Technologies, a Lucent/Bell Labs spin-off ( Flarion says that 3G mobile networks, although meant to carry voice and data traffic simultaneously, retain a circuit-switched, hierarchical architecture. Flarion asserts that the resulting design compromises of 3G networks, which are optimized for voice, impair their ability to deliver high-speed, low-latency data cost effectively. The company flatly states that the resulting high cost-per-MB of data delivery over 3G networks will prevent the emergence of mass-market wireless Internet access.

Flarion is developing a technology, called flash-OFDM™, which utilizes OFDM for mobile users and is directly compatible with the Internet. OFDM (orthogonal frequency division multiplexing) is based on a mathematical concept called FFT (fast Fourier transform) and has several implementation variants. The technology is becoming available in a variety of fixed wireless systems, mostly either in the unlicensed 2.4 GHz spectrum or in licensed LMDS/MMDS systems. However, the initial products Flarion Technologies is bringing to market are targeting the 700 MHz spectrum that the FCC designated for mobile wireless voice and data services. Flarion believes an orthogonal OFDM system can avoid interference that happens within a CDMA network with multiple users talking to one cell site. OFDM also offers a larger pipe than CDMA-based systems. Assuming that the spectral efficiency holds for mobile OFDM use, and latency times are good, then there may be no need for using WAP. Any interactive application could be just set up and run as is. Flarion says that its OFDM airlink enables 3 times higher data rate than a CDMA 3G airlink.

Another contender for XG dollars is Melbourne, FL Tantivy, an Intel funded company whose technology for portable systems is based on IP, IS-95, and CDMA. The meaning of tantivy is a rapid gallop or ride, and that’s what the company is offering its customers and users. The Tantivy Access Network link (TANlink™) I-CDMA™ technology supports simultaneous users transmitting and receiving data at rates as high as 368 Kbps. The company claims that its I-CDMA technology provides more than 10 times the Internet data capacity of other 2G and 3G network technologies at a fraction of the cost of deployment by minimizing dead air time and optimizing Internet data capacity. It also says its system effectively doubles the coverage area of the average wireless voice cell site, enabling more subscribers to be served with a single wireless base station. Using I-CDMA technology, Tantivy says wireless carriers can serve more than 1,000 users per sector, covering ranges as far as five miles.

Given the wild ride that 3G seems to be enjoying thus far, this is one horse race with no sure winners, so take your pick from any of the above. Just place your bets wisely, or you could lose your shirt -- or get it embarrassingly caught in your forecast zipper.

Copyright 2001 Francis Vale, All Rights Reserved

21st, The VXM Network,