Truly Illuminating

On the table in Thomas Little’s Photonics Center lab sits a disassembled traffic signal, light bulbs of all sorts, and two neon yellow flashlights that represent the future of wireless communication.

The flashlights, facing each other about six feet apart, use light-emitting-diodes, or LEDs, which need far less energy than incandescent or fluorescent bulbs. The LEDs have another advantage: they can be pulsed at high frequencies, transmitting the ones and zeros of data by turning on and off with a rapidity that cannot be detected by the human eye — but could be by a laptop, a cell phone, and countless other devices.

As LEDs become cheaper, they will increasingly replace traditional lighting sources. Little, a College of Engineering professor of electrical and computer engineering, and his colleagues in BU’s Smart Lighting Center are designing ways for these low-energy LEDs to not only illuminate our offices, homes, classrooms, and vehicles, but to network them as well. Their work is part of an $18.5 million, 10-year grant from the National Science Foundation awarded to BU, Rensselaer Polytechnic Institute, and the University of New Mexico.

The basic technology of LEDs has been around for a while. Traffic lights have used arrays of green, yellow, and red LEDS for years. While they need only about 15 percent of the electricity traditional bulbs use, LEDs haven’t been more widely adopted because researchers have only recently found a reliable way to produce white LED light and because the technology is expensive (an LED bulb can cost more than $80). Indeed, the two universities working with BU are developing new materials and components that will lead to more controllable and low-cost LED devices. Little says that with an estimated 22 percent of the world’s electricity used for lighting, “if the center does nothing else but help a transition to low-cost LED lighting, it will be an enormous gain for the world.”

It is the anticipation of a world lit by LEDs that makes the networking applications particularly exciting. “Imagine if network access were as ubiquitous as light,” says Little. “Wherever there is human-made light, you’ll potentially have access.”

One of the flashlights on Little’s lab table is hooked up to a laptop the researchers use to send a series of ones and zeros to a transmitter they’ve built into the flashlight to key its LED bulb on or off according to the data sequence. The facing flashlight has a receiver that picks up the transmission. Another prototype nearby can send and receive data in both directions.

Of course, Little and his team are working on grander projects than flashlights swapping ones and zeroes. They envision a world where overhead lights in an airplane or a classroom can transmit high-definition video, where a car’s brake lights can talk to another car’s headlights to initiate faster, automatic braking in an emergency, and where cell phones in the same building can be linked solely by the office light fixtures.

Each of those scenarios poses significant challenges to the BU researchers. How wide an angle of field should the light cover to be effective, for instance? Two neighboring passengers in an airplane may not want to watch the same movie, after all. And what about shadows? You might want your electronic devices to stay connected to a network even when something blocks the overhead light. Indeed, Little’s team will also investigate how to couple an LED network with WiFi or fiber networks to ensure a seamless range of connectivity.

Other hurdles will be creating standards and protocols for LED networking and increasing the data transmission speed so that it’s comparable with current broadband connections. Little’s flashlights operate in the kilobits-per-second range, a speed he likens to “an old-school dial-up modem.” One of his undergraduate students, Jimmy Chau (ENG’09), is working to push connection speeds into the one-to-10 megabit-per-second range, a project he hopes to complete by this spring.

Meanwhile, the researchers are investigating other means of supercharging LED data transmission, using, for example, the different wavelengths of LED light as individual data channels or amplifying the signal with multiple light sources.

Little notes that even if LED data networks aren’t able to immediately match the capacity of current broadband services, the speed for individual users could be comparable, because there would be far fewer people making use of any single connection. “WiFi networks advertise data rates such as 54 megabits per second, but practical speeds are much lower due to high traffic on the system — just like you experience driving on congested roadways,” he says. “With an LED lighting source, there will be many more inexpensive access points available to you.”

Little says that bridging the gap from the lab’s flashlights to commercially available LED data networks could take as few as five years. “Just think,” he says, “WiFi was still in the lab just 10 years ago, and now it’s everywhere. It’s remarkable how quickly these technologies can be adopted.”

Chris Berdik can be reached at cberdik@bu.edu.