Kevin Holley   


Kevin Holley,
European Telecommunications Standards Institute
Chairman, and Mobile Systems Design Manager, BT.


Judith Berck, Editor, GSM Data Today, February 16, 1998.

Kevin Holley has been involved in the development of the GSM standard since 1988, particularly on the data services. He is now Chairman of SMG4, the ETSI data development group for GSM and UMTS [Universal Mobile Telecommunications System], the "third generation" of cellular standards, which is expected to be implemented within five years.

While UMTS promises much higher rates of data transmission, we should not forget the current 9.6 bps rate already gives users capabilities for the applications they want most: corporate e-mail, Internet access, LAN/database access, fax, and short messaging service. There’s no need to wait – it’s here right now. UMTS will build on what we already can do today. Click here for more information on UMTS.

Much of Holley's discussion about new short-term technologies has to do with packet-switching. Current technologies are primarily circuit-switched, meaning a continuous circuit transmission allows the network to route continuous data to a single location. Circuit-switched data requires a dedicated radio channel even when no data is being sent. With packet-switched data, the computer that is connected to the cell phone sends and receives bursts, or packets, of data. A radio channel is occupied only for the duration of the data transmission instead of continuously, making it more efficient than circuit-switched.

Kevin Holley answered questions about what, when, why and wherefore the new UMTS standards, implementation obstacles and security. He also spoke about shorter term technologies: GPRS, packetization.

GDT: What will the new UMTS standards bring to corporations in regards to data that they can't get today?

KH: First is a greatly increased speed of data. UMTS aims to give up to two megabits per second data rate over the network, and that would be in a kind of closed environment, say in your office. The minimum that’s defined by UMTS is 144, 000 bps, which would be in an open area. [Editor’s note: Current data transmission speed is 9,600 bps.]

The other thing that it’s going to bring is more packetization, which means that if you make a call for say, three minutes, you won’t pay three minutes air time. You’ll pay for 500K on the downlink and about 80K on the uplink. So you only pay for what you transfer rather than paying for the time. So there will be a cost savings.

GDT: When do you expect UMTS to be implemented?

KH: UMTS people are talking about 2002, 2003. We’ve got a whole lot of things to go through like who’s going to get a license, what coverage are we going to offer, where are they going to start, because you won’t get UMTS just like that. So even though launch is about 2002, 2003, it could well be later before its really ready for a corporate to pick up and run with it.

For UMTS, what we’ve agreed is the radio interface will be based on that particular type of technology, but that really is the first stage. Having decided that you want that type of technology, you then need to decide exactly how you’re going to code it, how the model is going to work, how is the network going to work, and how does that thing connect through into wherever you’re trying to get to.

GDT: GPRS [General Packet Radio Services] is also packet-based, and expected to come much sooner than UMTS, towards the end of next year. Could you talk more about how packet-switched data is going to be an advantage over the current system?

KH: The primary advantage of packet-switched is that as the user you only pay for what you use as opposed to paying for a thirty minute call for only three or four minutes of airtime, for instance. There’s a secondary advantage, which is more in the operator’s direction than the user’s direction. And that’s if you have a lot of users all trying to use data at the same time, with packet data you can even it out so that everybody gets it fast. With circuit-switched, you’ve only got a certain number of channels, usually six or seven. If everyone is using that, then not only do they prevent other people from using data, but they also prevent other people from using voice. So an operator then will have to put in much more bandwidth to support a lot of data users if they’re using circuit-switched than if they’re using packet-switched.

I think it will also enable corporates to tailor what they’re doing much more easily, because they can design things like their own internal websites for both office and mobile. I know for example where I work, there is an internal webpage for doing travel claims. You fill the form out and then you hit the submit button and off it goes. But to do that over GSM now would be horrendous, because you have to spend half an hour while you’re typing it in, and most of the time you’re just reading and you’re not actually doing the data transfer. So this new packetization I think will be very useful for the corporate. That means that now rather than having to make two versions of things, one if you’re in the office and one if you’re offline, you can have one standardized way of doing things. I think that will make it a lot easier.

GDT: What do you see as the biggest obstacle to implementation of these new standards and technologies?

KH: Timing. Because we’ve got 14,400, high-speed, GPRS and UMTS all in the next five years, including the planning and trialing. So here we are in ’98, and we’re saying UMTS by 2003, that’s five years, and GPRS by the end of next year. I would say that the biggest obstacle is going to be getting it in that short space of time. It may well be some people will try to get into UMTS early and some other people won’t.

My big fear is that operators will be reluctant to introduce GPRS.

GDT: Why might they be reluctant to introduce GPRS?

KH: Well, operators tend to be steered by where they can get spectrum. UMTS gives them extra spectrum, but GPRS doesn’t give any extra spectrum. GPRS gives them an added value on their own spectrum, and it means they can make better use of their own spectrum. UMTS gives them an advantage in that they get some spectrum somewhere else. So the danger is that the operator then goes and focuses on UMTS because of the spectrum issue and then misses out on GPRS.

Now that might mean that there are no good developments of data services, because we’d still be at 9.6, 14.4, and it’s still circuit-switched data. I think UMTS really does need people to go out there and find applications that work well with data, and that I think can only work once we’ve got GPRS in. So I see GPRS as a springboard to UMTS in terms of data services. GPRS gives you the access, but it doesn’t give you the speed of UMTS. So I think we need to grow the market for data services first, before we start growing the speed of GPRS and UMTS.

GDT: How do you suggest that could happen?

KH: What we need to happen is for operators to make the step to GPRS. And to work much more together with corporates, to find out what are the key applications that they want to run, and to establish their links directly with the corporates.

And I think another issue is the one of roaming. Because at the moment, if you’re roaming and you want access to corporate networks, quite often you have to dial an international voice call and you’re paying an international voice rate. When instead you could be paying a local voice rate and then international data transport, because it’s more efficient to bundle the data links together than it is to use the voice circuits. There’s been some talk of even using the Internet for that.

GDT: What about security? Will security issues be handled differently with these new standards?

KH: We started out analog systems using sort of basic signaling, which anybody could pick up off the air, and the specs were available for decoding it, there’s no encryption of anything. That really was wide open to abuse.

With GSM, rather than going just a little bit more of a step, we went a whole big leap. We introduced an awful lot of security into the system right from the start. So GSM provides very effective security at the moment. Now we’d be kidding ourselves if we thought something couldn’t eventually happen. Obviously there is no problem at the moment, but UMTS will have even more advanced security.

GDT: Do you expect the market for data to take off once the UMTS standards are in place?

KH: Well, if you look at what’s happening on fixed networks, there is still, I believe it’s true to say, some growth in the fixed network voice market. There’s definitely big growth in the fixed network data market. You’ve got to look at why is that. People have been given something that they haven’t had before, access to the Internet -- which is a global information engine, if you like.

At the moment, if you get access to the Internet on the mobile phone, it is very difficult to do and it's very expensive. So if we can install an infrastructure such that you pay only for what you use, and you have the connectivity easily, then I believe that people will use the Internet. Now I don’t mean hunting around the Internet for something that interests you. I see the big use of data in the corporate sense for the INTRANET, where people are able to control information and are able to get information to people who need it very quickly.

GDT: What are the intellectual property rights issues that you anticipate regarding the third generation standards?

KH: That’s a good question. From my perspective, what is important is that we try to maintain the basis we’ve used in the past for IPR. That would be that most of the people involved advise ETSI if they do have IPR, and then hopefully they would make their IPR available to other people on reasonable terms. It don’t generally tend to go any further than saying reasonable terms, and then there are a lot of negotiations behind closed doors. The objective is to grow the pie as a whole, and hopefully not to restrict that particular technology you have to your own use. Because no one company in the world can service the needs of all the uses in the world.

GDT: What multimedia applications are you most excited about watching as the new standards are implemented?

KH: We always talk about multimedia because it kind of conjures up an exciting image in your mind. Obviously the new standards will help real-time video by supplying the speed that somebody needs.

GDT: Anything else you'd like to add concerning data and the new standards?

KH: I think the main thing that we haven’t discussed that will bring benefit for data users as well as any other users is the globalization of standards. The more global the standard is, the more manufacturers can concentrate on getting things right like the battery life, and they’re not focused on producing ten different phones for the different markets, so they can put their resources into making things work more effectively for the users.

Read Kevin Holley's Paper:

Evolution of GSM Data towards UMTS