The Future Foretold: Customer Demand Drives 3G

Looking into the future of next-generation technology, you may see the convergence of wireless and wireline worlds, a move from circuit-based to packet-based networks or more data-centric applications. But much of the future beyond 2G remains blurry.

Although you can't foresee what 3G enhancements will be, what changes will be necessary or how much they will cost, there is one certainty. Prepping your networktoday will determine your success tomorrow in meeting your customers' demands.

Future EnhancementsThe declared highlights of 3G are higher data speeds and increased capacity rates.

But it's easy to be skeptical when carriers have barely broken in 2G networks, and those were designed to handle more data traffic, anyway.

"Third generation is really about being able to merge voice, data and video," said Rick Rotondo, Excel Switching manager of industry marketing. "(End users) are going to see a lot of services being offered where, for example, (they'll) be able to get reasonably good Internet access speeds, something on the order of ISDN speeds."

In addition, he said that 3G would enable end users to have call waiting and to be able to flip back and forth between their Internet connections and voice connections. Their voice mailboxes would become multimedia mailboxes, possibly with video mail, but certainly e-mail, voice mail and fax all in the same mailbox.

Many carriers are preparing for these enhancements now. For example, Keith Paglusch, Sprint PCS senior vice president of technical services & network operations, said Sprint will add more functionality and features to its voice-mail system to make it more productive.

"The things that we're working on are things that work from a data perspective, clearly making this not only voice but data that you can use with your e-mail," he said.

Gerry Flynn, Bell Atlantic Mobile (BAM) senior member, technical staff, agreed there's an emerging demand for data services, and his company is working on its migration path to 64kb/s to 144kb/s services. But because data still is a small part of the total picture, determining what future enhancements to implement and what customer base to focus on is no easy task.

"The services that our customers want in the future or who our customers are in the future may be significantly different than our existing customers," Flynn said.

As carriers begin migrating to 3G, the next generation of switching will help dictate which applications come to fruition. When you're making equipment decisions, look at the flexibility of the equipment and the services it will allow you to add in the future.

Rod Nelson, AT&T Wireless senior vice president and CTO, said there will be a "radical departure" from the switches and networks that comprise wireless today. He envisions more distributed-network systems, and said the trend will be toward several different server platforms.

"The switch becomes more of a device that connects input to output," he said. "More value will go into software components that operate on a distributed server."

Sprint PCS is enhancing its switching functionality. According to Paglusch, because Sprint PCS' switching is fairly standard across all of its markets, it is adding intelligence outside of normal switching software in an off-board HLR. The carrier keeps customer databases in several signal-transfer-point HLRs across the country.

As Sprint PCS moves forward, it will add more features to its voice-mail system. A distributed architecture could offer other functions such as bundled services and improved call-processing capacity.

Because interoperability is so important, Rotondo said carriers already are embracing open models as they move through 2G toward 3G. The open model allows you to control your own destiny so you don't have to wait for your switching vendor to develop a feature. With open switching systems, you can try out new services, and if they don't work, you can reprogram the switch to do something else, Rotondo said.

"You can have an open distributed switching network and still keep it circuit switching, which is what everybody knows and loves," he explained. "There's a place for packet, but it is (the move from circuit to packet) going to be a transition as opposed to a revolution."

In the next five to 10 years, you may be delivering many different services, Flynn said. Because no one has a good sense of what exactly those services will be or what market will exist for them, it's difficult to prepare.

Flynn said an open, decentralized system would evolve BAM's networks so it can add 3G multimedia services and packet-based services without changing out its entire network.

"Our goals are to make sure that we have a flexible switching fabric in our networks so that we can evolve and introduce these services as the markets develop," he said.

Preparing Networks for 3G

All providers will have to update their networks and switches to stay competitive in the 3G world. Preparing your network for 3G enhancements will require early planning and working closely with vendors.

Based on its technology choices, AT&T's 3G implementation will be fairly straightforward and low-risk, Nelson said.

"In some cases, it will require software upgrades," he said. "Worst case, it will mean adding more channels at existing sites."

Paglusch said Sprint PCS is working to ensure that whatever it adds from a 3G perspective is backward compatible to its 2G products. It is advocating plug-and-play network upgrades rather than a forklift approach.

Many vendors are working on solutions to help ease the process. According to Dolores Grossmann, Lucent Technologies product manager for the 5ESS 2000 wireless application, 3G will require incremental additions and investments so carriers can leverage what they have today.

Rotondo said Excel is integrating and beta-testing its own voice-over-IP (VoIP) solution that allows providers who have used its switches in the past to add VoIP capabilities without really changing their applications.

Nortel Networks' switching of the future also will allow use of existing equipment.

Peter MacLaren, Nortel Networks vice president of strategic market development, wireless networks, said there already is a tremendous amount of capability that providers can evolve and enhance with complementary capabilities.

One solution, MacLaren said, is segmenting traffic before the switch, between circuit-switched voice and packet-data services, and moving the data services as close to the end user as possible onto packet backbone networks. These networks could be frame relay, ATM or IP networks. Providing an interface to the packet-based backbone network at the (base-station controller) level would enable more efficient implementation. The next step, he said, is to move more services to the packet/IP-based infrastructure as it is built up to optimize the cost of network evolution. This solution will be available within the next two years.

As IP-based technology becomes more cost-effective, you can grow your network based on that technology and continue to evolve legacy technology seamlessly, MacLaren said. The next generation of air interface standards will exist alongside current systems, the same way that current systems exist alongside analog. It's going to be an evolution with a lot of overlap and joint service with 2G radio technologies, MacLaren added.

"It (3G) might be a more cost-effective, higher-speed modem, or a better display of downloaded information," he added. "It's not going to be a technology discontinuity. It's going to be seen as another service or more cost-effective service."

If You Implement It, Will They Come?

Naysayers argue that the industry is rushing into 3G when it may not be necessary yet, especially when some providers still are paying for 2G infrastructure.

Vendors don't even agree on how much it will cost to implement 3G technology.

Rotondo said that making a complete change from 2G to 3G would be difficult, particularly for GSM because it will have to go to a CDMA format. Rotondo doesn't expect that will be backward compatible.

"I haven't seen anybody make the argument yet that (3G) is going to be cost-effective," Rotondo said.

Other vendors said 3G might not cost as much as 2G because providers shouldn't have to replace entire base stations. Lucent and Nortel are working on solutions that will use existing infrastructure in deploying 3G capabilities.

"Our philosophy is evolution so that we protect our customers' investment," Lucent's Grossmann said.

Because vendors can't predict what 3G investment will be required, most carriers are waiting for evidence of customer demand before they invest too much time and money in 3G switches.

"Like any other type of investment we make, it would be supported by a business case that says there's a market opportunity," BAM's Flynn said.

BAM, which plans 3G domestic field trials with Lucent later this year, will start with technical trials before doing market trials with customers.

"We're approaching this as something we're going to do in evolutionary stages, so we will deploy it where a market emerges and start out to maximize the utilization of our existing frequencies and spectrum plans," he said. "We're not talking about a deployment where we're going to change all of our network infrastructure overnight, because I expect that the devices and the market interest would be something that evolves over a matter of several months, if not years."

According to Flynn, the market is shifting, and the proportion and characteristics of BAM's digital customers are growing rapidly. He said emerging market opportunities for data are going to hit a lot faster than you think, "but we haven't really seen that slope come in the form of real customer demand that says, 'I need 144kb.'"

But carriers agree it's just a matter of time. According to Nelson, AT&T's corporate customers are asking for wideband capabilities now, and he is confident that customer demand for 3G will grow. AT&T Wireless' customers already have expressed interest in wireless data, higher data speeds, residential cable modems and features such as an integrated mailbox and the ability to check e-mail via voice, said Ken Woo, director of corporate communications.

"We have to ask ourselves, 'Are we providing 3kb/s to 4kb/s per data bit,' and if so, we're going to be successful with customers," Nelson said.

Sprint PCS is monitoring customer interest closely as well.

"If it isn't something that our customers want, then we're not going to do it," Paglusch said.

He said 3G wouldn't attract just corporate subscribers. If 3G delivers the higher data rates and additional capacity it promises, then there could be an equal mix of user interest including business users as well as consumers with high-speed requirements.

Although Paglusch is not sure what those customers will pay for 3G enhancements, he explained that Sprint PCS' potential investment would require a high payoff.

"I'm not going to go put in 3G and not be able to make a return on that investment," he said. "We will use the same kind of criteria to upgrade to 3G that we did to get into the business -- that it has to be cost-effective for us to upgrade to 3G, and in that equation, we have to have a revenue or subscriber base that supports it."

Sprint PCS is planning a lab trial in 2000 that will last six to 12 months. It will make its 3G decision based on its findings.

"We are very, very active in the standards process for 3G because we want to make sure that what we're going to buy later meets our needs," Paglusch said. "But we're not out there ready to put in 3G until there's such a time that customer demand or our own capacity requirements make that necessary."

The best thing to do now is to stop trying to predict the future and start taking steps to plan for it. Although no one is exactly sure how or when, 3G is coming, and your networks need to be ready.