Antenna Maker ArrayComm Pursues Wireless ISP Plans


Antenna-technology developer ArrayComm Inc. is working with
unnamed partners to develop a nationwide high-speed-data-access system.

The move signals that previous assumptions about the
transport capacity of fixed wireless networks will soon be overturned.

Starting late next year, ArrayComm expects large-scale
field trials of its "i-BURST" system to begin in preparation for the launch of a
1-megabit-per-second Internet-access service in 100 cities representing 60 percent of the
U.S. population, chairman and CEO Martin Cooper said.

"For this system to be real, it has to be ubiquitous,
and we think this is the scale we have to start at," he added.

Cooper said the anticipated timing of commercial rollout of
the system -- slated for sometime in early 2002 -- would put ArrayComm and its carrier and
vendor partners in position to capture a large share of the anticipated mass market for
high-speed wireless services.

"We're not competing with wireline services,
because our system is designed to give users access wherever they are," he added.

ArrayComm expects the cost of deployment to be under $2
billion. Officials said there's a second phase to the plan that will raise the data
rate to about 2 mbps.

ArrayComm's move -- rooted in its proprietary
spatial-division multiple-access antenna technology -- comes as a number of entities,
including AT&T Corp., are tapping smart-antenna and other techniques to enable them to
deliver high-speed services over relatively limited amounts of spectrum.

If they use small segments available below the 2-gigahertz
level that are not already used for mobile voice, broadcast TV and other applications,
they can reach users with saturation coverage, both inside and outside of their homes and
offices, because the lower frequencies have superior propagation characteristics.

ArrayComm needs the Federal Communications Commission to
license a sliver of spectrum on the order of 5 megahertz to 10 MHz wide, Cooper said. The
company's technology supports transmissions at a rate of 4 bits per hertz or
frequency cycle, which translates into 40 mbps if 10 MHz of spectrum is used, he added.

By focusing a portion of the RF energy on each connection
that is sufficient to deliver 1 mbps in both directions, only in each instance where the
communication is delivering a signal, thousands of users can be served over a tiny segment
of spectrum from each base station transmitter/receiver.

"The technology allows you to talk to more than one
person at a time on the RF channel in the same time slot," Cooper said.

That translates into spectral efficiency that is 400 times
that of current-generation cellular and 40 times that of third-generation digital
personal-communications systems that are slated to hit the market in 2002, Cooper added

"Remember," he said, "we're giving
something up, which is the need to provide mobile service, so we can do things that these
systems can't do."

Already supporting commercial services in Japan, Malaysia,
the United Arab Emirates and, soon, China, ArrayComm has a head start over other antenna
makers. But other suppliers are gearing up to provide systems that will make portable
high-speed wireless-access services a reality within two years.

Engineers at Lucent Technologies' Bell Laboratories
are testing cutting-edge antenna technologies in various deployment configurations in and
around Crawford Hill, N.J., which offer systems integrators a variety of options for
addressing wireless-capacity-expansion needs, said Rich Howard, director of Lucent's
wireless lab.

"We've come a long way in figuring out how to
integrate and apply our innovations in real-world networking situations," Howard
said. Now, he added, it's a question of when market demand will drive systems
integrators to begin making use of the technology.

"If you believe there's going to be a widespread
market push for wireless data, wireless operators will have no choice but to use every
trick in the book to improve bandwidth efficiency," Howard said. "What
we're working on is the biggest trick of all."

Lucent's wireless lab group is confident that the
advanced antenna techniques could be quickly put to use in network systems to effect
10-fold to 20-fold increases in capacity over a given wireless link and to overcome the
interference problems that will intensify as the use of wireless spectrum for delivering
services in competition with wireline networks accelerates.

These techniques include an innovation announced late last
year, known as "BLAST." That approach uses multiple transmit and receive
antennas to exploit the multipath nature of wireless communications, and a marriage of
this technique with steerable antenna-beam technology in a multipurpose combination can
address market conditions ranging from urban centers to the suburban fringes.

BLAST employs Lucent-developed algorithms to assign
specific signals to specific transmission paths in the multipath dissemination of a radio
wave at a given frequency, thereby allowing reuse of the frequency many times over for
delivery of different messages to and from different users.

In contrast, the company's multibeam-antenna system
serves multiple customers over the same frequency by moving a wide beam from customer to
customer in quick, millisecond hops that are timed to coincide with the time slots
assigned to each customer in a time-division multiplexing configuration.

BLAST is ideal for an urban environment, where densely
packed building surfaces create the reflective patterns needed for multipath
communications. The steerable-beam technology is good for increasing the
information-carrying capacity in suburban areas.

By employing multiple-antenna arrays at the transmit and
receive ends, Lucent has found that it can combine the two techniques to maximize the
benefits of each, depending on where a base station is located, Howard said.

"The hardware is identical -- multiple antennas with
multiple radios," he added. The key is to activate the hardware via algorithms that
match the local situation.

Today, digital-signal processors are still too expensive to
allow for multiple antennas at the receive end, Howard said. But within the next year or
so, the cost curve could fall far enough to make commercial applications of the integrated
advanced antenna system feasible.

Nobody is more interested in deriving higher data-carrying
capacity from limited spectrum than AT&T Wireless. That unit's executives have
spoken often about "Project Angel" as a wireless solution to delivering
high-speed-data and voice services over cellular, PCS and other spectrum.