re: EDGE Odds & Ends by Ian Channing
Close to Seeing the Light of Day ...
>> Close To The Edge?
Ian Channing
Mobile Communications International
Issue 90
01 April 2002
EDGE recently re-entered the limelight following the decision by several major US networks to go down the GSM/GPRS/EDGE route to 3G. Now the spotlight has moved to Europe where GSM operators are said to be contemplating deploying EDGE as an adjunct to WCDMA. Is this a realistic scenario? Ian Channing looks at EDGE's chances in the GSM market.
It has to be said that EDGE has a rather chequered history. Originally conceived as a key step in the evolution from GSM to 3G, EDGE was muscled out as GPRS and 3G got crowded into the same space. So then it was reinvented as the technology of choice for GSM operators that failed to win 3G licences.
As the majority of GSM operators did, in fact, win 3G licences, there seemed, at the time, to be no purpose in them deploying EDGE. The technology was next cast as the solution for North and South American TDMA operators looking to deliver 3G-style services. A special version of EDGE - TDMA EDGE - was developed to meet the requirements of these operators and there was, at least for a while, considerable enthusiasm for the approach.
However, the decision by AT&T Wireless Services and other major TDMA operators to convert their networks to GSM/GPRS and to evolve to EDGE via that path, effectively scuppered TDMA EDGE. Now the focus has come full circle, back to the concept of GSM operators deploying EDGE as an adjunct to their WCDMA networks.
Given the pressure on resources faced by operators and vendors, this seems an unlikely possibility. But there are some heavyweight players pushing the notion. At the 3GSM World Congress in Cannes, Siemens hosted a meeting of the EDGE Operators Forum, which was attended by 130 operators and vendors including AT&T Wireless, Cingular, Motorola, Nortel, Nokia and Ericsson. In a statement following the meeting the group reaffirmed their faith in EDGE, stating that, "some 90 per cent of operators worldwide [are] expected to commit to WCDMA and EDGE strategies en route to 3G." With all due respect to the enthusiasm of the EDGE camp, this actually says very little.
There is no doubt that the commitment to make EDGE happen in North America exists. In addition to AT&T Wireless Services, Cingular Wireless, Microcell, VoiceStream, Rogers of Canada and Mexico's Telcel are all committed to following the path through GSM and GPRS to EDGE. Contracts worth several billion dollars have already been granted and the first GSM/GPRS services from these new operators should be coming on stream this year, with EDGE services following 12 to 18 months later.
In addition, there have been hints that one of the largest North American non-GSM operators - Verizon - is contemplating moving away from its planned IS-95/CDMA200 1x evolutionary path and embracing GSM/GPRS/EDGE. In a little noted statement made around 12 months ago, a middle level Verizon executive averred that although the company was deploying CDMA 1x, it was also looking at the GSM evolutionary path.
This seems a little unlikely, though. Despite behind-the-scenes pressure from Vodafone, Verizon's official CDMA line has not wavered and Vodafone's recently reported discussions with Qualcomm hint at a chipset roaming solution to the transatlantic problem. On the other hand, never say never...
The potential prospects for EDGE in Latin America are less clear than in the US. Even before the most recent developments north of the border, Latin American TDMA operators had exhibited limited enthusiasm for the TDMA EDGE route. The CDMA lobby has invested considerable time and effort in trying to persuade the Latin American operators to select CDMA2000 1x as their upgrade path. Mark Whitton, vice president and general manager, GSM Americas, Wireless Networks Nortel, observes Latin American TDMA operators splitting into two camps. "One group are probably going to convert to GSM and it makes sense for them to deploy EDGE-capable base stations in their new networks. This will give them the ability to introduce EDGE anytime they want. Other TDMA operators will probably choose the CDMA 1x route." Alan Hadden of the Global Mobile Suppliers Association points out that the decision by US, Canadian and Mexican TDMA operators to go GSM/GPRS means that of the 90 million TDMA subscribers worldwide, 80 million are now in the GSM camp.
With the heavyweight operators supporting it, EDGE seems certain to take off in North America (handset supply notwithstanding), but what are the realistic chances of GSM operators, particularly those in Europe, deploying EDGE? And what benefits would they gain if they did so? Most importantly, what would deploying EDGE cost the operators?
In theory it should be easy to determine whether European GSM operators are planning to deploy EDGE, but in reality this is not the case. Contacting a random selection of major GSM operators proved a fruitless exercise - not even a "no comment" was forthcoming. Even the GSM Association, usually eager to lay claim to all aspects of GSM and its evolutions, was uncharacteristically retiring.
One can understand the reticence of the operators and their representatives. They have paid enormous sums of money - over $100 billion across Europe - for 3G licences, and seen their stock market valuation tumble as a consequence. Against this backdrop it would surely be difficult to admit that 3G was going to be considerably delayed and that, as a result, they were going to deploy an alternative - and largely unknown - technology in the interim.
After all the time and marketing effort spent talking up the benefits of 3G, retreating from this position, albeit for the best of reasons, would hardly be viewed with equanimity by the markets. One industry participant, who declined to be named, believes that European GSM operators are unwilling to talk up EDGE for fear of detracting from WCDMA.
What is certain is that no contracts for EDGE equipment have yet been announced in Europe. But as major vendors are at pains to point out, EDGE is a 'simple' upgrade rather than a new network and as such would probably be covered by the ongoing frame contracts which all operators have in place with their vendors.
Nokia says it has 13 EDGE trials ongoing, although the Finnish vendor is unable to name the participants. Torsten Hunte, strategic product manager for data at Ericsson says that there has been, and continues to be, much discussion with European operators about EDGE.
Tim Penhale-Jones, marketing director of Sony Semiconductors and Devices in Europe, says he is quietly confident that EDGE will pick up over the next year or so in Europe as well as in the US. "My personal view is that if 3G slips much more the drawbacks of GPRS and the advantages of EDGE will come through. With relatively little investment operators can generate a lot more revenue by having EDGE in the interim and maybe enjoy three more years of additional revenue, that GPRS won't bring them, before 3G comes along."
Assuming that GSM operators are contemplating deploying EDGE, are all the elements in place to allow them do so - and in what time scale? For obvious reasons, infrastructure vendors are bullish on the prospects for EDGE and are doing their best to help operators along this road. Nokia says that all its base station products will be EDGE compliant within months - all new introductions have had EDGE for the last year or so. Ericsson claims that its base stations have all been EDGE compliant since 1995 and Nortel says that it has been shipping EDGE-capable base stations since 1997.
According to the vendors, if the operator already has EDGE-compliant base stations then the upgrade is simply a matter of installing EDGE transceivers and making a software upgrade. "For some operators, especially those who are building their networks for the first time like the Americans, EDGE is simply a software upgrade," says Harry Kuosa, EDGE marketing manager at Nokia Networks. "If you are still building coverage, as many operators are in Asia and Africa, there is no cost because you are just putting in new hardware. On top of that, all operators are constantly improving capacity in their networks by putting in more transceivers." In that situation, says Kuosa, the operator has the choice of putting in GSM-only transceivers or GSM/EDGE transceivers. Given that the cost is roughly similar, why not put in GSM/EDGE? This view underpins the constantly reiterated pitch that EDGE is essentially a low-price option. However, for mature GSM operators with very large networks, Vodafone or Deutsche Telekom for example, it's a different story. Mature networks can contain thousands of base stations and it is inevitable that a proportion of these will not be EDGE-enabled.
In the worst-case scenario this could mean that an operator upgrading his network to EDGE would have to physically change a whole cellsite and, as Nortel's Whitton points out, "you do not want to get involved in forklifting in new base stations."
Although the EDGE lobby would have operators believe that this technology is a low-cost option, the reality appears to be that it cannot be costed accurately except in the case of greenfield deployments. For European and Asian operators of long standing, EDGE expenditure looks somewhat open-ended. Mark Paxman, principal consultant in the Wireless Technology Group at PA Consulting, compares EDGE investment with that required for GPRS. "The beautiful thing about GPRS is that you didn't really need to change the base stations. You needed the software load and an upgrade to the core network but there was very little to do at the base station.
With EDGE you probably need a new base station or you need new transceiver cards in the BTS. You need to buy a lot of kit and you need to roll a truck out to every cell site and plug it in and test it, so it is more expensive." As a basis for comparison it should be noted that UK consultancy Ovum once priced the upgrade for GPRS in a mature network at $150-200 million.
Originally, EDGE represented one step in an elegant progression which went from GSM to GPRS then to EDGE and finally to 3G. With technological compression having put paid to that concept, the question remains: if a GSM operator does plan to deploy EDGE, how will it fit into his overall strategy, particularly in relationship to the ongoing implementation of 3G? The approach advocated by EDGE proponents is that it should be deployed as a complement to WCDMA. The thinking is that operators will initially deploy WCDMA in major cities but the low density of potential customers will inhibit them from putting WCDMA into rural areas. Enter EDGE. By deploying EDGE in rural areas operators will be able to offer, hopefully, seamless high speed data services on a nationwide basis - albeit using two different technologies.
"EDGE will allow operators to deploy '3G-like' services in rural areas in a very cost effective way," says Ericsson's Hunte. "They already have the GSM infrastructure nationwide so going for EDGE is a minor investment. They see EDGE as a complement to their WCDMA coverage because although they will deploy wideband in the major cities, no-one will start doing it in the country." Although this approach has a certain logic there are regulatory issues.
Most European operators have 3G licences which mandate them to deploy WCDMA to cover a certain percentage of the population within a certain timeframe. Depending on the country's demographics this requirement could be met by installing WCDMA in the top 20 cities, possibly allowing the operator to deploy EDGE in rural areas without caveat from the regulator.
However, in some instances, Sweden and Spain for example, the licence conditions are more draconian and demand both population and area coverage, making it unlikely that the regulator would overlook the substitution of EDGE for WCDMA in rural areas. In any event, as EDGE is classed as a 3G technology, will regulators see its deployment as being in breach of licence conditions? If EDGE deployment does move to the top of European operator's agendas one can see some delicate negotiations ahead.
Given that deploying EDGE will involve unquantifiable capital expenditure, what will the technology offer operators? Vendors claim an increase in capacity of at least threefold, an important benefit given that most GSM operators are pushing the limits of their spectrum allocations. There should also be improvements in quality. "It's a win-win situation for operators," says Hunte. "They'll be gaining capacity where it is needed today and at the same time improving the packet data performance from GPRS to such an extent that they can run the same services on EDGE as on WCDMA."
EDGE will also deliver higher data speeds than GPRS although the precise level of improvement is contentious. The most widely quoted figure for EDGE data speed is 384kbps - which was the basis for the claim that EDGE is a 3G technology. In fact, vendors have been revising this figure upwards of late and are now claiming 59.2kbps per timeslot, a total of 473kbps if all eight timeslots on a GSM frame are utilised. Although no-one is prepared to admit to it, this is unrealistic at best. Unfortunately, constant repetition has made this the 'headline figure' beloved by the popular press and, as with GPRS, it will adversely affect end-user expectations.
A more realistic consensus is that EDGE will deliver around three times the best average speed of GPRS - 30-40kbps - which will equate to 100kbps plus in good conditions. The issue of achievable data speeds also impacts on the deployment question. EDGE is being proposed as the solution for high-speed data in areas not covered by WCDMA. It is reasonable to assume that operators will initially deploy EDGE on their 900MHz networks as these provide nationwide coverage.
The question is, what impact on coverage will there be if high-speed data is being delivered over this network? PA Consulting's Paxman says the impact will be negative. "Because of the way that EDGE works, EDGE cells for high data rates are very, very small, probably of similar size to WCDMA cells. EDGE is quite progressive but as you wind up to the higher data rates you get smaller cells than standard GSM 900 unless you start putting smart antennas and that sort of thing on top. Basically, if I want to get 140kbps over EDGE it will only cover some fraction of the cell over which I would normally get GSM 900 voice."
Nokia Networks' Kuosa disagrees, saying that improvements to the standard have overcome early problems. "With EDGE you will get more capacity, better quality and higher data rates. You will also get better coverage, from simulations the incremental gain is averaging 3dB. For existing operators this means you will get better indoor coverage in the cities and better coverage in rural areas.
The incremental gain was not there in the early stages of the standard, only the new modulation, and if you only have this you have a bigger bit pipe but the distance will be less. With the incremental redundancy you get better coverage."
As experience with technologies such as HSCSD and GPRS has shown, the key to success is handset availability. "From an RF standpoint the addition of EDGE is not that challenging," comments Sony's Penhale-Jones, "We are going for a single chip solution that does GSM, GPRS and EDGE. In GPRS some of the software issues held up the handsets but in terms of the evolution to EDGE it is not a huge amount of effort. It is certainly nowhere near the complexity of going to 3G. We have samples now and there will be production by the end of this year. For a handset that you are releasing in 2003, vendors can design the product in and be confident that they will have an EDGE compatible handset for next year."
Fresh from a recent meeting of the EDGE Operator's Forum, the GSA's Hadden was happy to confirm that EDGE handsets would be available: "By 2004 several of the handset vendors are saying that EDGE will be standard in all devices and the leading players say that by then they will have a wide portfolio of products supporting GSM, GPRS, WCDMA and EDGE." That EDGE will be an established fact in North America is looking most likely. Almost certainly a significant number of South American TDMA operators will choose the GSM/GPRS/EDGE route to satisfy their 3G needs. With regard to Europe, until one of the GSM operators breaks ranks the jury is still out.
According to Hadden, quite a few European operators are committing to EDGE and much will be revealed over the next weeks and months.
One suspects that concerns over possible negative reaction from shareholders, markets and regulators are the main reason for the continued silence.
There are also likely to be, in the current climate, problems with resources given that operators and vendors are struggling to get WCDMA off the ground whilst still supporting ongoing GPRS and even GSM deployments. Had the 3G runaway train been kept in check, EDGE would have figured more prominently in the development of operators' technological evolution.
It is possible that GSM operators will slot it back into their plans, but how simple this will be and whether it will prove the right decision in the longer term is open to question. <<
- Eric - |
