Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Disclaimer: The Parliamentary Library does not warrant or accept liability for the accuracy or usefulness of the transcripts. These are copied directly from the broadcaster's website.
Li-Fi: the LED-based alternative to household -

View in ParlViewView other Segments

[Audio: phone ringing]

Robyn Williams: That's a telephone from the old days. You may recall that the man behind the phone was Alexander Graham Bell, the man from Edinburgh who went to Canada. He not only invented the phone he also produced the photo phone, an idea which led now in 2013 to light carrying the internet and other systems of communication.

This Science Show on RN begins in Edinburgh, and I'm actually standing in Alexander Graham Bell House with Professor Harald Haas whose team has developed this revolutionary device, which promises that lights communicating with devices will give us intelligent buildings, crash-free cars and any number of other remarkable innovations. Harald, how was it done?

Harald Haas: We have taken white LEDs, light emitting diodes, which now form the new basis for energy-efficient lighting, and we have changed the intensity of that light in relation to the information that needs to be transmitted, and the intensity changes will then be received by a photo detector which is in your mobile phone or in your mobile devices, and the changes of intensity will then be converted back into datastream.

Robyn Williams: So you have to do something to an LED, to a light, to make it have that capacity, you can't just use any LED?

Harald Haas: You could also use an incandescent light bulb, but the transmission rate would be in the number of bits per second, whereas we transmit gigabits per second. So it's quite a big difference in terms of what you would be able to do with a normal light bulb.

Robyn Williams: How does the light carry that information?

Harald Haas: The light carries the information in the changes of the intensity. So you get, for example, when you want to transmit a binary one you would turn on the light, if you want to transmit a binary zero you would turn it off. And if you do that very rapidly a human eye wouldn't be able to see whether the light is on or off, it would basically have a constant brightness, but still you would be transmitting a very high number of bits per time.

Robyn Williams: Enough for the internet?

Harald Haas: Far more than you could now achieve with your 3G system. We have shown now three gigabits per second via an off-the-shelf LED light bulb.

Robyn Williams: Let's go through to your lab. Is it far away?

Harald Haas: It is about a minute walk.

Robyn Williams: You're from Germany, are you?

Harald Haas: I am from Germany, yes indeed.

Robyn Williams: But you did your qualifications here?

Harald Haas: I did my PhD here 16 years ago, yes. I had an appointment at a private university in Germany, and I'm back in Edinburgh since 2007.

Robyn Williams: I must say it's quite extraordinary that light should be the transmitter…we're just going through the hallway here…because the Australians from CSIRO famously developed wi-fi when they'd been searching for black holes, and the techniques that they used for that sort of astronomical search word then adapted, being used for wi-fi for laptops all over the world. Is your technique in any way a rival for wi-fi?

Harald Haas: No, our technique is a compliment to wi-fi. So wi-fi has its limitation.

Robyn Williams: What are the limitations?

Harald Haas: That it requires electromagnetic spectrum, radio frequency spectrum, but radio frequency spectrum is limited, it's crowded, we have many systems that use the RF spectrum, military systems, broadcast systems. The bandwidth of interest, which is between 0 and 10 gigahertz is very crowded and there's hardly any free space, and while wi-fi is brilliant technology, it suffers from the lack of spectrum, and we have plenty of spectrum. Light spectrum is 10,000 times larger than the entire radio frequency spectrum, and we basically leverage this free available resource.

Robyn Williams: Here's your lab, and you've got one, two, three, four, five, six, seven, eight people working away.

Harald Haas: If I can show you at the back here we have a running horse, we transmit a sequence of images or a video through this little blue LED light bulb here. Where we have the LED here, the transmitter there, and we transmit this fast running horse here.

Robyn Williams: Hi, I'm Robyn.

Hamid: Hi, I'm Hamid.

Robyn Williams: In front of a screen, and there's the horse going like crazy, running away, and that has been just by a little light, that's quite amazing.

Harald Haas: That's only a little light, it's a very small micro-LED, and if you can imagine that you'd take any of these, for example 100 LEDs that form this micro-LED and transmit separate datastreams like this, it's incredible what could be done in terms of data rates. And with this one LED, for example, we have achieved three gigabits per second.

Robyn Williams: That's immensely powerful, from simple light transmissions. So you could piggyback on light and cross the world, but do you have to be able to see the destination? Do you have to be in visual touch with it?

Harald Haas: No. People always believe you need to have line of sight, so you need to see the light and you need to see the receiver and you need to have a direct connection between both elements, but in fact it would also work if you block the transmitter and would hold the receiver against the wall. The reflected light from a wall would be sufficient photons that would enable the transmission of a video like this. And we have so-called digital modulation techniques that would adapt to the amount of signal that comes into the receiver.

Robyn Williams: This means that you could almost wire an entire building and make it interactive.

Harald Haas: You could wire up an entire building. Every light source you could imagine, these are ceiling lights, wall lights, flashlights of mobile phones, headlights of cars, all these could be potential sources of high-speed wireless transmitters. Imagine what this could enable, this could enable very smart systems. For example, going back to the home where we know that 30% of the entire energy consumption is in domestic environments, and the reason is because we have inefficient heating systems, we have inefficient use of energy in homes. But if you could develop clever systems that would turn on the heating when people are in a room, then you could save a lot of energy. You need intelligent systems, you need intelligent devices that communicate wirelessly in order to explore these capabilities to reduce energy consumption in the home. Yes, you can make homes smarter, it's the so-called internet of things or machine-to-machine communication that would enable the next move into new innovations.

Robyn Williams: But presumably driving the system would then have to be energy efficient because if it costs you a gigantic amount to run with all this lighting, then it would be counter-productive.

Harald Haas: This is a classical misconception…

Robyn Williams: I'm full of them!

Harald Haas: No, no, no, it's for me to explain that. People believe that the light has to be on, like full brightness, to communicate wirelessly with light but this is not the case. We can turn off the light to levels that it appears to be off to the human eye but it would still have the capability to act as a transmission device.

Robyn Williams: And what about other lights that were being turned on and off in between, even daylight? Would you not to get an interfering effect?

Harald Haas: This is another very good question. Daylight is constant light. The way we encode data is by changing the intensity according to the data we want to transmit. If we know that interference, which is the sunlight, is constant light we can filter that out, and then the receiver would only see the differences in the intensity and it would link that to the datastream.

Robyn Williams: You wanted to take me to your office where your company is set up.

Harald Haas: Yes, I can show you here another experiment we do in the space of multiple input, multiple output. I've told you we have one LED here but you could also take a multiple LEDs, and now what we have seen here with one stream you can now have in parallel, we can have many parallel streams, and every stream could transmit one gigabit. If you have, for example, RGB, red-green-blue, and every colour would transmit three gigabit per second and you run RGB together which makes white, you can transmit nine gigabits per second, because you would transmit on each colour channel. And this is shown here on the oscilloscope. And this receiver would make sense, you would separate all these different colours and then would enhance the data rate even further.

Robyn Williams: Sorry to interrupt your work.

Stefan: Hi, my name is Stefan.

Robyn Williams: Where from?

Stefan: I'm from Bulgaria.

Robyn Williams: You're from Bulgaria! Here I am in Edinburgh and I'm yet to meet a Scots person.

Stefan: Well, we do have some Scottish people in our team, English as well, but it's a very diverse multinational team.

Robyn Williams: I'm sure it would be quite unknown to most people, using light to run things like the internet. Do you find when you talk to your friends and colleagues that they don't know much about this?

Stefan: As it happens, most of my friends and colleagues are either scientists or engineers, so a lot of them are open to the idea. But yes, most other people who are not in the field of science or engineering, yes, it's something completely new that people are yet to discover for themselves.

Robyn Williams: So now we go across to what might be your commercial outfit. It just so happens that in Australia, there is huge discussion about having a national broadband network. Is all that going to be out of date pretty soon?

Harald Haas: No, it will not be out of date, and we still need the high-speed backbone, but we had have an exponential increase of data volume through our mobile networks, and in the last 50 years the only technical innovation that has created transmission capability was the concept of smaller cells. In the third generation broadband or wireless networks we had very large so-called radio cells. In the third generation the cells were shrunk from 35-kilometre radius to 5-kilometre radius and down to 1-kilometre radius. In LTE, Long Term Evolution, we talk about 100-metre cell radius, so you can imagine how many base stations you would need to put up in a city like this, like Edinburgh. And on the other hand, if you take the concept of a light bulb, a light bulb has a very confined area it could illuminate. We can take this concept of smaller cells even further, and the increase of the…

Robyn Williams: We're just going in the building, out of the wind.

Harald Haas: And even if you look at radio cells outside and you want to communicate, for example, in an elevator like this, the radio waves would be attenuated, this is a Faraday cage here, there is no RF signal here, but look here, there is light. So you could even communicate in buildings, in basements when there is no radiofrequency reception but light would be there. It would be a sort of optical cell.

Robyn Williams: We are in a lift, obviously as you can hear, and it's metal-clad, so you'd get no signal here but you can with your system through the light because we've got light on.

Harald Haas: Yes, we could have started a video transmission downstairs and we had uninterrupted video transmission through this elevator here.

Robyn Williams: Incredible.

Harald Haas: And it would be handed over to this light system here.

Robyn Williams: So as long as you've got light you have a kind of linking connection. The fact that there are brick walls in the way and darkness wouldn't matter.

Harald Haas: It wouldn't matter because it would cover these gaps. One gigabit per second as we move here, the system would hand over to this ceiling light here.

Robyn Williams: Here's an office.

Harald Haas: And these are all the employees here, a couple of people here you see, ex PhD students.

Robyn Williams: Hi, I'm Robyn Williams from Australia, ABC. Where are you from?

Man: I am from Bangladesh…

Robyn Williams: You're from Bangladesh. Any Scots here? No? Where are you from?

Man: I'm from Macedonia.

Man: And I'm from Spain.

Robyn Williams: Spain!

Harald Haas: But we have one here.

Robyn Williams: Are you Scottish?

Man: No, north-east England.

Robyn Williams: You come from north-east England!

Harald Haas: So we have seen the research lab, now we see this module here which would go into a ceiling or would go on a desk, and you have a ceiling light like the one there. So now we see the three videos running in parallel, that is all transmitted from that light source there.

Robyn Williams: I'm seeing a person in a garden, I'm seeing a flower blooming, and I'm seeing a person at home. It's just three different videos all running on one screen all coming off that light. Being theoretical, if you can have transmission from wherever there is light, what about under water?

Harald Haas: It's a very good example. Radiofrequency signals wouldn't propagate underwater because water is salty, it's conductive and therefore no propagation possible. Light on the other hand, if you take remote operated vehicles they need light in order to illuminate the sea ground to see what's there, and these lights could also be used to transmit data from one underwater unit vehicle to another one. Yes, that is a classic example.

The other example is intrinsically safe environments like petrochemical plants where you can't use RF because of sparks and so on, light could be used and light is plentiful there. In ships where there is lots of metal which shields RF but there is, again, lots of light. In hospitals where you have, for example, MRI scanners, MRIs produce a lot of electromagnetic interference which would not work with RF, but light would be there. So doctors could see the image that they scanned directly on their hand-held in the room where they take the photograph. It enables a whole lot of new applications.

And going further with a car, car-to-car communication, car lights are made of LEDs in the front, in the back, so you could envisage systems that talk, car-to-car communications, talk to each other, and prevent collisions by the system reacting faster than the human can react.

Robyn Williams: That sounds wonderful. By RF presumably you mean radio frequency?

Harald Haas: Radio frequency, that's correct.

Robyn Williams: Okay, so this is a commercial setup. Is it really an existing commercial operation? Are you in business?

Harald Haas: We are in business, and this unit that you see here is in production. We have a range of customers, we call it beta customers, with whom we do the first experiments, yes, it's all up and running.

Robyn Williams: And it's really impressive to watch, I must say, those tiny LEDs transmitting like powerhouses. And when you consider that, say, on YouTube 72 hours of videos are uploaded every 50 minutes, we are going to need networks with greater capacity to cope. Professor Harald Haas with me at Alexander Graham Bell House, University of Edinburgh.