The Rutgers University Winlab
By Michael Grebb, Wed May 21 00:00:00 GMT 2003

The Wireless Information Network Laboratory finds itself at the center of an industry increasingly eager to capitalize on university research.


Once upon a time, research and development (R&D) departments were the backbone of technology companies, fueling innovation and commerce. But since those heady days, corporations have found it more efficient (and cheaper) to shift much of that R&D into higher education where enthusiastic students and professors are willing to stay up all night figuring out better mousetraps. Without the usual overhead.

Wireless firms have funded several university programs that aim to develop new technologies that make better use of spectrum capacity, frequency efficiency, and battery power, among other areas. More rare are programs focused solely on wireless Internet technologies, which have acquired increased importance as wireless-enabled laptops and personal digital assistants (PDAs) proliferate. For such narrow focus on the mobile Web, corporations are increasingly turning to one place: WINLAB.

WINLAB, or the Wireless Information Network Laboratory, was founded at New Jersey's Rutgers University in 1989 to develop wireless networking technology. As the Internet exploded over the last decade, the program tackled everything from mobile computing to high-speed modem design to radio resource management to network architectures and protocols. More than 20 global corporate sponsors from the U.S., Europe, and Japan now pay up to $70,000 annually for the right to license intellectual property produced by WINLAB's research (Associate members with limited research access pay $30,000 per year).

"It's a very interesting shift," says Dipankar Raychaudhuri, who became WINLAB's director in July 2001. "Corporations are becoming more short-term driven. Their R&D labs are starting to shrink. Universities are more efficient. The raw human talent is there." The program has about 20 faculty and staff, along with a fluctuating core of 30 to 50 students.

So far, WINLAB has produced some 75 graduates who obtained Masters or Doctorate degrees in wireless technology and have gone to work in the wireless sector. The center also receives grants from the National Science Foundation (NSF), the Defense Advanced Research Projects Agency (DARPA) and the New Jersey Commission on Science and Technology (NJCST). So like many similar academic programs, WINLAB feeds off a support axis made up of academia, private industry, and government.

Infostations: Drive-ByData?


One of WINLAB's most intriguing projects in the new age of third-generation (3G) networks is known simply as "Infostations." The idea is to create ultra high-speed, short-range local area networks (LANs) that could blast massive amounts of data to mobile devices in quick bursts lasting only moments, depending on the content. The ultimate goal is to develop base stations capable of 500 megabits per second, although current technology maxes out around 54 Mbps. "We have a good chance of reaching 500," says Raychaudhuri. "It's very real. It's just a few years away." The concept becomes most interesting when applied as a supplement to future 3G networks that support "always-on" connections. Here's how it would work:

A pedestrian or motorist with a wireless PDA would pass by an Infostation (basically a souped-up wireless LAN base station), which could detect the device and immediately dump huge streams of data into its hard drive. Assuming WINLAB researchers eventually reach the 500-Mbps goal, someone could theoretically download the text of several newspapers and maybe even a few video clips in less than one second. Infostations could even be constantly updated from a remote central server to ensure the most recent information gets blasted to mobile devices.

"An Infostation is like a wireless cache," says Raychaudhuri. "Popular content gets pushed to an Infostation and then gets pushed to consumers in an opportunistic fashion. You could even put personalized information in an Infostation." Under that scenario, a person could get their customized news, stock quotes, sports scores, and other data updated automatically to their wireless device every time they pass by an Infostation. It would not only detect the device itself, but that a particular user owned the device.

Indeed, such base stations could be strategically placed in downtown areas-constantly shooting data to any wireless device that even momentarily entered its range of up to 300 feet-and relieve stress on the cellular network at large by offloading the highest bandwidth usage. As WINLABers are fond of noting, this creates a variation on the "anytime-anywhere" coverage of traditional cellular systems. With Infostations, as they like to say, it's "manytime-manywhere." In addition, WINLAB researchers claim that Infostation deployment across 3G networks could reduce by two or three times the buildout costs because they are cheaper to deploy than cellular towers.

WINLAB researchers envision several uses for Infostations: Museums could send tour information to people as they pass through the front door. Airports could beam flight information to devices in the terminals. A retail store could send detailed product descriptions to shoppers. Travelers could even receive specific data on every town they enter, listing maps and even pointing out local landmarks, gas stations, and restaurants. "Pretty much any neighborhood you're in, you could have information on it," says Richard Frenkiel, WINLAB's director of strategic planning. "As you pass from Infostation to Infostation, it could update you."

Multi-modal


In WINLAB's universe, such services would work just as well for motorists as they do for pedestrians. Services could shoot data such as maps, navigation tools, or even video-on-demand to motorists either lost or just seeking to silence their kids in the back seat. "You could watch movies as if they are broadcast to you, when in fact it's being downloaded in chunks as you pass by Infostations," says Ivan Seskar, WINLAB's associate director of information technology. In other words, a car barreling down the highway would grab five or ten minutes of a movie from an Infostation every few miles, creating the illusion to the viewer that it was being continuously streamed.

Of course, such ambitions may be considerably far off. "To make it successful, you'd have to be on every highway," notes Christine Loredo, a senior analyst at the Strategis Group. "And that wouldn't be cheap." She adds that such services would be limited to the U.S. and Canada, considering that Europeans and Asians don't share the same "road-trip" culture prevalent in North America.

And this all assumes that the integration of Infostations and larger 3G networks would actually work. So far, it's all just theory. And most experts agree that for such a synergy to effectively make efficient use of bandwidth, the consumer must not trip over the complexities of the network. Wireless devices must be able to sense incoming and outgoing data both from the 3G network as well as the vicinity of the Infostation without any hiccups that would require the user to constantly switch device settings. "I see how it could offload some capacity," says Loredo, "but it's got to be seamless. The consumer can't know about it." Agrees Raychaudhuri: "You have to keep it hidden from the consumer. If you do, then it just acts like a turbo boost to your service."

Eventually, WINLAB researchers theorize that Infostations could even democratize access to wireless networks. At roughly $200 per Infostation, Frenkiel says even the smallest businesses could afford to deploy them in their stores, helping them compete more ably with the Wal-Marts of the world. "One of the many ideas is a mom-and-pop Infostation," notes Frenkiel. "When you're near a grocery store or something, you can access anything you want. It would seem like anyone could put these in. But I'm not really sure that's how it's going to play out." So how will it? "That's the 64-thousand dollar question," he says. "It's hard to predict what people will use this for."

How to Build It, and WillThey Come?


And therein lies the rub. After all, no grocery store is going to install Infostations in the aisles unless a good number of people walking in the front door are armed with always-on wireless devices. No content producers are going to create automatic downloading services or on on-the-go video on demand unless Infostations are relatively ubiquitous. That's years away. And that chicken-and-egg problem isn't even worth discussing until 3G deployment becomes a widespread reality-a prospect not helped by the constant barrage of deployment delays around the world, especially in the U.S. Even assuming 3G deployments accelerate, it's unclear what level of broadband content users will demand via their wireless devices.

There's a big difference between news text or stock quotes or video-on-demand movies. And will enough people use such broadband services to require the use of supplemental Infostations? "All of this really depends on when these 3G networks take off," says Loredo.

Other hurdles also exist. For one thing, wireless LANs that use the 802.11 protocol use unlicensed bands that are prone to interference. Such problems become even trickier when moving pedestrians or motorists are constantly coming in and out of range of each base station. Loredo notes that trees on highways and tall buildings in cities could prove quite vexing to Infostation reception, requiring potentially much trial-and-error in figuring out where to position each one to get maximum coverage.

But WINLAB researchers note that regulators could eventually allocate specific spectrum for Infostation usage. And because Infostations need such a short range of signal reach, extremely high frequency bands (which are unsuitable for cellular networks because signals degrade very quickly) would be adequate.

Despite all of the uncertainties, WINLAB researchers are undeterred. Raychaudhuri notes that other WINLAB projects are addressing many outlying issues, including one new program (currently shoring up funding) that will research ways to perfect "system-on-a-chip" architectures that promote low power consumption, smaller devices, and "always-on" architectures. Raychaudhuri even foresees a day in which small system-on-a-chip badges on clothing could sense body heat and turn up the air conditioning in the room when needed. "We have to apply Moore's Law to wireless," he says.

The researchers at WINLAB appear to be giving it their best shot.

It's Mobile Research Week on TheFeature! Check back daily for reports, analysis and in-depth articles about the wireless research community.

Michael Grebb has previously written for The 'now-defunct' Industry Standard, Business 2.0, and eCompany. From Washington DC, he covers the impact of mobile technology on modern society.