AT&T Corp., among other players on the wireless front,
is rapidly reaching the point where it will be able to deliver high-speed-data and voice
services over its personal-communications-services and cellular spectrum.
Lately, most discussions about AT&T's local
telephone play have been centered on its bid to acquire Tele-Communications Inc., which
would give AT&T a coaxial doorway into homes.
But AT&T executives were adamant that wireless remains
a crucial component of the company's strategy to attain a ubiquitous presence for
what it calls "all-band services." More specific, the packetization of voice,
data and video communications into the Internet-protocol format supports AT&T's
hope for an "always-on" provisioning of bandwidth on-demand.
"All of our products, services and applications are
going to come with all-band, always-on functionality built in," said Dan Shulman,
president of AT&T Worldnet, the company's data-services arm.
Where wireless is concerned, AT&T is less than two
years away from achieving what it calls "3G," or third-generation data capacity,
over its existing infrastructure, said Doug Brandon, vice president for external affairs
at AT&T Wireless Services.
"We'll be able to support at least 384 kilobits
per second [over mobile connections]," he said. That capability -- including speeds
in excess of 1 megabit per second over fixed links -- will require allocation of only 2
megahertz of the available spectrum, he added.
AT&T's emphasis on integrated IP services over all
of its access platforms comes at a time when other wireless players are beginning to
recognize that their traditional thinking about data misses the true implications of the
emerging 3G high-speed capabilities.
By wrapping voice in an all-IP framework, wireless
operators could directly compete with the integrated-service strategies of the wireline
carriers -- but with the added advantage of having a mobile component, said Merle Gilmore,
president of the communications-enterprises unit at Motorola Inc.
Wireless phone operators should focus on "the
transition away from circuit-switch-dominated, hierarchical architectures to a
distributed-IP packet-switched architecture," Gilmore said.
"[Doing that] will facilitate a single architecture,
capable of delivering the voice services and the data applications, whether you're
using GSM [Global System for Mobile Communications], CDMA [code-division multiple-access],
or any of the new standards that are being proposed," Gilmore added.
A growing number of options, while confusing the
standards-setting process, are opening a near-term window for wireless operators to move
into high-speed-data services, officials said.
Those could include fixed, integrated IP services that
compete directly with the types of advanced services envisioned in the cable
industry's PacketCable initiative.
Qualcomm Inc., for example, recently took the wireless
industry by surprise with its introduction of a new system that would allow CDMA-based
carriers to introduce high-speed data through simple chip-set upgrades of existing
Qualcomm's HDR (High Data Rate) technology would allow
mobile-service operators to deliver packet communications at up to 2.4 mbps, said Irwin
Jacobs, chairman and CEO of Qualcomm.
"It's like an Ethernet, but without collision,
which means that you can support a lot of users," Jacobs added.
Unlike 3G systems, which are designed to support voice and
data access over a 5-MHz channel, the HDR system requires only 1.25 MHz of bandwidth for
data-only passage of one or more of the "IS-95," or "CDMAOne,"
channels. Thus, HDR is a better way to maximize data efficiency, Jacobs said.
"With data, you can burst out packets at very high
rates and pass around access [to time slots] among multiple users," he added.
Data over PCS and cellular today, when it's offered at
all, typically runs at 14.4 kbps. New systems slated to enter the market in the time frame
set for HDR top out at 144 kbps.
Qualcomm expects carriers to begin testing the new system
next year, and to roll out services commercially by the end of 1999, officials said.
However, commercial acceptance will require support from handset manufacturers, which may
be loathe to invest in a proprietary iteration of CDMA.
That reluctance is likely because of several standards
initiatives, both domestic and international, which are already generating heated debate
over how the industry can evolve to higher data speeds.
"This is too new a development to say where it
leads," said David Murashige, vice president of CDMA marketing and product management
at Nortel, a licensee of Qualcomm's technology. "One of the nice things about
Qualcomm is that they are pushing the envelope, but this complicates an already-difficult
discussion about where the wireless industry is going with data."
With AT&T setting a benchmark at 384 kbps for mobile
data and higher rates for fixed -- coupled with Qualcomm coming in with a 1-mbps-plus
capability for mobile and fixed services -- competitive pressures are building on carriers
to act sooner, rather than later.
But, as often happens with technology, new techniques are
coming into conflict with the slower pace of standards, with the net effect of stasis.
One example of this is the CDMAOne camp, which worked in
consort to push a standard known as "IS-95B" to completion earlier this year.
Already, that group appears to be losing interest in the IS-95B system, which, among other
things, supports data access over widely deployed CDMA infrastructures at 64 kbps without
any hardware upgrades. Now, with so many other options capturing operators'
attention, vendors are starting to back away from introducing IS-95B systems.
"There's still a lot of uncertainty about what
the right solution is going to be, including whether the whole idea of 3G is the right
approach," Murashige said.
At this point, only one infrastructure vendor -- Motorola
-- is supporting IS-95B, which had been targeted as the evolutionary path for CDMA. Now,
the CDMA community appears to be shifting support to another, more bandwidth-efficient
standard, known as both "IXRTT" and "CDMA2000 Basic."
That approach will deliver data simultaneously to four to
five users over a 1.25-MHz channel, versus delivery to only two users, as is the case with
IXRTT is still in flux, but it could uptick to 144 kbps
throughput, rather than 64 kbps, by the time the draft specification is completed early
It will require some hardware upgrades of existing
infrastructure components, meaning that wireless carriers will need to change out ASICs
(application-specific integrated circuits) in about "20 percent of what we have on
the ground today," Murashige said.