This article was taken from the February 2012 issue of Wired magazine. Be the first to read Wired's articles in print before they're posted online, and get your hands on loads of additional content by subscribing online.
Using off-the-shelf electronics, he can stream videos using an ordinary light bulb fitted with signal-processing technology of his own design. The lamp shines directly on to a hole cut into the oblong box on which it sits. Inside this box is a receiver that converts the light signal into a high-speed data stream, and a transmitter that projects the data on to a screen as a short video.
If Haas puts his hand in front of the lamp, excluding the light, the video stops.
Haas, 43, holds the chair of mobile communications at Edinburgh University's Institute for Digital Communications. His demo is scientifically groundbreaking: it proves that large amounts of data, in multiple parallel streams, can be transferred using various forms of light (infrared, ultraviolet and visible). The technology, he says, has huge commercial potential. His device can be used with regular lighting and electronics -- albeit reconfigured -- and could transform the way we access everything from video to games, accelerating the speed of internet access by many hundreds of megabits. It could let us download movies from the lamps in our homes, read maps from streetlights and listen to music from illuminated billboards in the street.
Haas's discovery is based on a subset of optical technology called visible light communication (VLC), or Li-Fi, as it has been dubbed. VLC exploits a hack of human perception: light-emitting diodes can be switched on and off faster than the naked eye can detect, causing the light source to appear to be on continuously.
Rapid on-off keying enables data transmission using binary code: switching on an LED is a logical "1", switching it off is a logical "0". Thereby flows the data.
The potential applications are enormous: divers working at depths could use light to communicate; air passengers could connect to the internet through the LEDs inside the aircraft. Haas sees the technology potentially disrupting industries from telecoms to advertising.
Research into VLC has been conducted in earnest since 2003, mainly in the UK, US, Germany, Korea and Japan. Experiments have shown that LEDs can be electronically adapted to transmit data wirelessly as well as to provide light. VLC is faster, safer and cheaper than other forms of wireless internet, advocates say -- and so could eliminate the need for costly mobile-phone radio masts.
German-born Haas sits in his office in Edinburgh University.
Several of the university's faculties, led by the informatics department, are involved in his project, called D-LIGHT; the university, a sponsor, owns the intellectual property. A month before wired's visit, Haas prompted stunned applause when he demonstrated his technology live to the audience at TEDGlobal in his adopted city. "To my disgrace," he confesses, "I didn't know anything about TED before [my talk]. Now it's like winning the lottery -- every day, I get five to ten emails, all very positive and excited. They are excited because it's easy to comprehend the unique opportunity in this technology."
In person, Haas is calm, humorous and cerebral, unfussed by his academic celebrity and unmotivated, he says, by the prospect of wealth. Married to a fellow German and graduate of the University of St Andrews, he jokes about their young children picking up an Edinburgh burr. He does not seek fame, he says, only for his work to fulfil its potential. "I'm less interested in what may change the world in 20 years' time," he says. "I'm looking into fundamentals of science that could do so within the next few years."
Haas grew up in Nuremberg. His thinking was greatly influenced by his father, an engineer at Siemens. He recalls a teenage fascination with LEDs -- a new technology at the time -- and inventing, at around 13, a rev counter that neighbours wanted to install in their cars. "I played a lot with electronics when I was a kid," he says. "I wanted to understand the transistor, and I wrote an essay [about them] at school. Everyone thought, 'What is this strange guy doing here?'"
He followed his father into Siemens, developing mobile-phone chips in Munich, but grew bored and escaped to academia. In 1995 he went on a scholarship to Mumbai, where he helped Siemens to develop a mobile-phone network. By 1999, he was at Edinburgh University, where he and his tutor, Gordon Povey, now his business partner, developed a patent for the 3G Universal Mobile Telecommunications System, which they sold to Siemens.
His office is now in the university's Alexander Graham Bell building. In 1880, Bell became the first person to send a voice message by modulated light. Using his "Photophone", Bell directed sunlight at a parabolic mirror that captured and projected his vocal vibrations. Sunlight's unpredictability sidelined his idea.
Radio transmission seemed more promising.
Now Haas is fulfilling Bell's ambition with LEDs, whose principal technology was discovered more than 100 years ago. LEDs were little used until the late 20th century, when they began appearing in car-instrument panels, infrared handsets and TVs. They are brighter, tougher and more reliable -- if costlier -- than the incandescent bulbs they are gradually replacing.
Haas has a small lab stuffed with equipment, including the now-famoustable lamp and its box of electronics. It was here in 2007 that his research assistant, Mostafa Afgani, first sent data using light signals. Haas's invention centres on how these signals are modulated: the information, embedded within visible light emitted from the LEDs, is transmitted by means of many subtle changes made to the intensity of the light at the ultra-high rate of 100 million cycles per second (100MHz). The photo-detector in Haas's box monitors these tiny variations and converts them back into a digital signal, from which the transmitted information is extracted.
Early in his research, Haas became intrigued by the properties of two wireless-technology systems: multiple-input and multiple-output (Mimo), which employs antennas at both the receiver and transmitter to improve signal; and orthogonal frequency-division multiplexing (OFDM), found in DAB radio and 4G mobile communications.
A weakness of OFDM is its peak-to-average power ratio, which produces wide fluctuations in the signal. But Haas could work this into an advantage for optical communications if he used those variations to crank up intensity. This meant that he could transmit data streams in parallel, and at rates limited only by the number of LEDs. He calls his transmission technique spatial modulation OFDM, or SIM-OFDM. "We exploit three dimensions," he explains. "Time, frequency and space. No one has ever done this before."
Haas, Povey and Afgani founded a company, VLC Ltd, to market their transceiver technology and a seed-funding round was completed last November. A larger "Series A" VC funding round is expected later this year. But the team is in a race to market: rival companies are already selling and trialling VLC products. In Tokyo, Nakagawa Laboratories, where Haas spent a semester in 2003, has developed what it calls underwater visible light communications: a means of picking up a scuba-diver's voice using a microphone in an LED light and sent to another diver. In December 2010, US technology firm LVX started providing light- powered broadband services to six public buildings in the Minnesotan town of St Cloud. LVX says it's demonstrating "fibre optics without the fibre". The company would not comment, but is said to have transmitted data at an unimpressive 3Mbps.
A VLC industry is beginning to cohere. Last autumn a report by industry researcher Global Business Intelligence, Visible Light Communication (VLC) -- A Potential Solution to the Global Wireless Spectrum Shortage, mentioned half a dozen companies that are active in the field, including Casio and Intel. Samsung and Boeing were not named, but are also thought to be exploring VLC. Last October, a group of companies from Germany, Norway, Israel and the US formed the Li-Fi Consortium to promote optical wireless communications.
It's easy to see why VLC could disrupt the way data is delivered. It offers relatively high bandwidth -- and visible light is free and safe. Imagine being able to use your laptop or tablet wherever in the world you can find a modified LED light. No more scrambling for access codes, or searching for hotspots, or risking leaks and electromagnetic interference, because artificial light, unlike radio waves, doesn't penetrate walls. You can control its beam, like a torch; and even if that torchlight is dimmed to the point of appearing switched off, you remain connected, as the LED signal is still operating.
Evangelists emphasise VLC's advantages in hazardous conditions or tough environments. They say it will transform air travel by allowing overhead cabin lights to connect mobiles and laptops in-flight; and that it will improve conditions for those working underwater -- on oil rigs, for example -- where radio waves cannot penetrate. Traffic lights could better regulate traffic flow using data; and LED car lights might alert drivers when other vehicles are too close. More pervasively, anyone seeking an internet connection could connect using light from streetlamps and electric shop signs.
One major downside is that artificial light cannot approach the range and penetration of radio waves: think of the limitations of a torch beam or car headlamp. So a mobile phone enabled by Li-Fi (through its inbuilt photo-receiver) will never be as useful as radio in the open air. Still, Li-Fi could certainly help relieve pressure on the existing networks' capacity. And it will offer an attractive alternative to the congested radio frequency spectrum.
The US Congress is contemplating whether to reallocate television broadcasters' unused spectrum and auction it for wireless broadband.
Li-Fi use in aeroplanes is a priority for Haas and Povey. One of the "ten or 11" applications they are considering for their technology could be spurred on by a deal with a German aircraft manufacturer they will not name. The German government has awarded them a €700,000 (£600,000) grant to develop a system that will provide continuous internet access via an aircraft's avionics data-bus system. Haas, though, has bigger ambitions: he wants to help the manufacturer build planes in modules. This would require less hard-wiring. "They want greener and lighter aircraft, and for seating to be flexible," he says. "And making things modular means you can't wire them up."
They are also in talks with the ClinicalResearch Imaging Centre at Queen's Medical Research Institute in Edinburgh. Radiology equipment is highly sensitive to electromagnetic interference, and the distribution of data produced by computerised axial tomography (CAT) or magnetic resonance imaging (MRI) scans requires extensive cabling. VLC would make it possible to distribute scans widely and quickly. But first the team must trial their Li-Fi prototype at a suitable location. They like the idea of a school: some local authorities have expressed concern about alleged health effect on pupils from radio transmitters and Wi-Fi.
Meanwhile, data transfer using VLC is getting faster and more efficient. Last year, researchers from Siemens Corporate Technology in Munich transmitted data over five metres with an Osram LED. And in August, the Heinrich Hertz Institute in Berlin claimed to have demonstrated 800Mbps transceivers under laboratory conditions; by comparison, few wireless service providers muster speeds above 100Mbps.
At TEDGlobal, Haas demonstrated a data rate of transmission of around 10Mbps -- comparable to a fairly good UK broadband connection. Two months later he had achieved 123Mbps. Now he's targeting 600Mbps. He has, effectively, thrown down a challenge to the competition: we can squeeze more data from a light bulb than you can.
Mark Leeson, a professor at Warwick University's School of Engineering, is more cautious about VLC's prospects. He told New Scientist last July that "there has been a lot of early hype".
While conceding to wired the "important contribution of the Edinburgh group", he thinks problems with the technology are yet to be solved: "The really big challenge is to provide an uplink from the receiving device back to the ceiling fixture. This may entail the use of a different wavelength in the infrared, or even RF
[radio frequency]. As currently presented, VLC is [only] a broadcast technology."
A common criticism is that any blocking or movement affecting the light source could interrupt the line-of-sight transmission -- as Haas demonstrated at TED. He dismisses this objection by gesturing to the space under his office table. "In a room, this will never happen," he says. "See. Even under a table, it gets a signal, because light travels in straight lines but is reflected easily." He points at the ceiling fixture. "If you have enough illumination, sufficient for reading, then you can guarantee enough signal power for wireless communication."
Haas argues that the transmission rate at the Hertz Institute "was not in real time". "They wouldn't be able to transmit a video in the way we demonstrated it at TED. In a comparable environment like ours, in fact, they achieve 83Mbps."
There's also the infrastructure challenge of getting data to the lights; after all, a light fixture has to be plugged into something. "Are people really going to wire network cabling to every single light bulb?" asks Dan Ryan, cofounder of ByteLight in Boston, whose VLC technology is focused on location applications. "There's been a lot of talk about using powerline communications to get data to the bulbs. But I have yet to see anyone reliably demonstrate this at high data rates." Povey answers this by minimising its importance. "Every wireless-communications system has a backhaul of some type," he says. "Many smart lighting systems already have control bus cables, which we can make redundant. Wi-Fi has the same issue: you need to run a cable to every access point.
Powerline communications are getting quite mature, and many people find them more reliable than using Wi-Fi. Wi-Fi speeds will not get anywhere near 600Mbps in a practical setting."
Even if his own technology is adopted, however, Ryan anticipates resistance to VLC from mobile-device manufacturers and the lighting industry. "Why would lighting manufacturers make their bulbs VLC-enabled if mobile devices can't receive the data? And why would a mobile-device manufacturer make their phones VLC-enabled if there are no bulbs to talk with? What we have is a chicken-and-egg problem. What will likely happen is that VLC will roll out in
'closed-system use' cases, where the facility manager controls both the lighting and the devices receiving the data. Military, hospital and factory-type environments all come to mind." "There are always detractors," Povey says. "Always people saying, 'You won't achieve these data rates.' In my previous companies, people kept on, 'Oh, you should give up now,' but I always succeeded in every one that I did."
VLC Ltd is currently looking for more investment, and seeking an initial partnership with a smart-lighting company to launch its first products by June 2012. Then, within a decade, Povey believes, the likes of Homebase and B&Q will be stocking their modified light bulbs.
But first, they must show their fully functioning VLC-enabled light bulb to a potential manufacturer. So they have enlisted the design expertise of a leader in LED lighting, the West Midlands company Integrated System Technologies, based in Aldridge. "We've demonstrated the technology and we've got some funding," says Haas. "But we shouldn't get overexcited. We are now in base camp. Walking further up the hill is another story."
Last October, Haas was a finalist at the prestigious World Technology Network Awards in New York City. The awards are presented to individuals and companies who, in the opinion of judges such as inventor Ray Kurzweil, are performing innovations of "the greatest likely long-term significance". Competing against the likes of Bell Labs, Yahoo! and Facebook, Haas lost to telecoms giant Alcatel-Lucent. "I think the nomination came a year too early," he shrugged after the awards ceremony. "After all, we don't have an actual product as yet."
Michael Watts wrote about OPOWER in 08.11
This article was originally published by WIRED UK
