Wireless is more
VCs are finally seeing the potential of
broadband wireless equipment.
By Alex Gove
Red Herring magazine
From the August 1999 issue
In the last two years, venture capitalists have shifted
their focus in the wireless communications industry.
After investing in startups that purchased and resold
the spectrum used by PCS and millimeter wave, VCs
are now concentrating on companies that are
developing high-speed data-delivery equipment for this
spectrum.
Because it was either unavailable
or grossly expensive, VCs were
until recently wary of investing in
wireless equipment used to
transport broadband data traffic.
But with significant advances in
capacity and reliability, along with
lower prices, wireless data
technology is finally looking like a
strong broadband alternative to
digital subscriber line (DSL) and cable modems.
WAVE GOOD BUY
The hurdle for broadband data delivery has long been
the last mile, or the distance between the telephone
company's central office and the home or business.
Cable modems and DSL can deliver data at speeds of
1 Mbps or higher, but each technology has its
limitations: DSL, for example, cannot currently scale
much beyond 6 Mbps, and wiring businesses for cable
is costly.
According to Kevin Maroni of
Spectrum Equity Investors, "Over
the next two years, the big action
in wireless lies in figuring out a
way to deliver broadband data at
what used to be considered
narrowband frequencies."
Nevertheless, using PCS and
millimeter-wave spectrum does
pose its own set of technological
and economic challenges.
Engineers have trouble cramming
thousands of bits of data into
PCS spectrum, which does not have much excess
capacity, while millimeter wave's high equipment costs
and line-of-sight limitations have slowed that
technology's adoption.
Mr. Maroni has mixed feelings about the current crop
of startups that are trying to provide broadband
wireless data to residential markets. Last September,
Cisco Systems (Nasdaq: CSCO) paid $157 million for
Clarity Wireless, a startup that claims to have solved
millimeter wave's line-of-sight problem. This
achievement would be especially beneficial in
residential markets, where the transmission path is
often obstructed by trees and other obstacles. But
Clarity's approach is new, and experts think that
delivering millimeter-wave data to homes will take a lot
of engineering.
As for PCS, Mr. Maroni does not believe that the
spectrum can deliver data to homes at speeds of more
than 2 Mbps. Because cable modems and DSL can
scale beyond this rate, and because the demand for
bandwidth will likely only continue to increase, he
compares investing in wireless broadband PCS to
"throwing money down a rat hole."
ROAM ALONE
Dave Hathaway of Venrock Associates sees more
commercial potential in broadband PCS. His firm has
invested in Tantivy, a startup in Melbourne, Florida,
that is developing equipment to plug into PCS carriers'
existing central offices and enable them to offer speeds
of up to 1 Mbps. Although third-generation, or 3G,
technology, which will likely incorporate some aspects
of the GSM standard, is supposed to deliver data at
speeds of 2 Mbps or higher, GSM was founded on
the premise that users should be able to roam between
cells. Tantivy's chip sets will not allow this; Mr.
Hathaway says that such roaming capability is
expensive and imposes heavy demands on a carrier's
infrastructure. He is confident that Tantivy-equipped
carriers can offer users 1-Mbps speed at a
substantially lower price than that of 3G technology.
The company expects to ship its product next year.
Venrock, New Enterprise Associates, North Bridge
Venture Partners, and Novak Biddle Venture Partners
have invested $6 million in Tantivy in two rounds of
financing. The company is in the process of raising a
third round of $7 million to $8 million.
Like Tantivy, Gigabit Wireless, a ten-person startup in
Menlo Park, California, is also trying to cram a large
amount of data into existing spectrum. But Gigabit
intends to offer the same benefits to both consumers
and businesses over low Multichannel Multipoint
Distribution Service (MMDS) frequencies. CEO
Arogyaswami Paulraj says that millimeter-wave
spectrum below 3 GHz is attractive for two reasons:
compared with higher MMDS frequencies -- like 22,
28, or 38 GHz -- it has fewer problems with foliage
interference, and the radio-frequency equipment is
cheaper. The trick is to increase the efficiency of this
spectrum, which is scarce and pricey. Mr. Paulraj
claims that Gigabit uses multiple spatial channels in the
same way that wavelength division multiplexing in
fiber-optic transmission employs different wavelengths
to push multiple data channels. Accel Partners has
invested $2.4 million in the company in one round of
financing.
Andy Fillat of Advent International says that all of his
companies are using more aggressive modulation
schemes to push more data into narrowband
frequencies, but these efforts have resulted in high
signal-to-noise ratios and shorter transmission
distances. Mr. Fillat, too, is interested in making better
use of lower millimeter-wave spectrum. He says that
newly opened NII bands at 5.3 and 5.7 GHz, as well
as lower frequencies like 2.4 GHz, are "getting good
reception among investors." He points in particular to
Wireless Incorporated, which makes digital microwave
radios for this space. But according to Mr. Fillat, the
high millimeter-wave frequencies have not caught on as
quickly as expected because investors overpaid for the
spectrum or the equipment, or both -- and "the
equipment isn't there," or up to speed yet. Although he
acknowledges that the wireless industry is on the verge
of a new generation of broadband radios, he thinks the
market will need a few years to digest this. "Right now,
I'm not sure what to look for in a new investment in
this area," he says.
NO SOAP RADIO
Mr. Fillat may question where the greatest investment
opportunity in the spectrum lies, but other investors are
much more bullish about how broadband radios will
affect broadband wireless data delivery at higher
millimeter-wave frequencies. Columbia Capital was a
leader in putting together two of the four broadband
wireless players -- WPN Communications (which was
sold to Nextlink [Nasdaq: NXLK] last year for $695
million) and Advanced Radio Telecom (Nasdaq:
ARTT). Although Columbia has made no recent
hardware investments, Managing Director Jim Fleming
says it is taking a hard look at this area. He adds that
current equipment can deliver data 15 to 25 percent
more cheaply than fiber, and he predicts that these
costs will continue to drop.
The reason that Mr. Fleming and others are so excited
about millimeter-wave data delivery is due largely to
the technological advances of a small startup in
Orlando, Florida. As we noted in this column in July
1998, Triton Network Systems has been developing
millimeter-wave radios that can push OC3 (155
Mbps) and OC12 (622 Mbps) traffic to corporate
campuses. The difference between this year and last is
that Triton is now delivering on its vision: James Wei of
Worldview Technology Partners says that Triton's
radios have achieved 99.999 percent reliability in field
tests and that these radios will be dramatically cheaper
and faster than a millimeter-wave carrier's current
options, including the T3 radios that P-Com (Nasdaq:
PCMS) now sells for $18,000 each. According to Mr.
Fillat, the price of the next-generation radios and hubs
could be $10,000 or less.
Because the economics of the industry are changing
rapidly, Mr. Wei and Bandel Carano of Oak
Investment Partners are exploring the idea of starting a
new millimeter-wave communications carrier in either
the United States or Asia.
BROADBAND-AID
This idea is not original; companies like WNP,
Advanced Radio Telecom, Teligent (Nasdaq: TGNT),
and WinStar (Nasdaq: WCII) have all been involved
in it. But Mr. Wei says that Triton's radios represent a
"massive dislocation" in terms of the money it takes to
build a wireless broadband carrier. Given the examples
of NorthPoint Communications (Nasdaq: NPNT) and
Rhythms NetConnections (Nasdaq: RTHM) -- two
DSL companies that recently went public, and
currently have valuations of $5 billion and $3.9 billion,
respectively -- he thinks the market will reward a
wireless broadband carrier that takes full advantage of
the next-generation equipment. And although Formus
Communications, one of Red Herring's top 100
companies of 1999 (see June 1999), has been pushing
this business model in South America and Europe, Mr.
Wei thinks that Asia is still ripe for the picking thanks
to its underdeveloped telecom infrastructure and the
high cost of laying fiber there.
Mr. Wei's thesis is interesting but risky. Many
companies have had trouble balancing the huge cost of
acquiring spectrum with the expense of deploying
equipment. Prices for spectrum are still high. Although
Columbia's Mark Warner, who has been investing in
wireless spectrum since 1982, remembers when the
markets complained about companies that paid $6 to
$7 per point of presence, spectrum will never be
anywhere near that cheap again. But with the expected
sharp drop in the price of wireless broadband
equipment, Mr. Wei's optimism seems appropriate --
especially now that, as Mr. Fleming says, "wireless
broadband technology is finally catching up with
reality." |
