HD video demonstrated streaming over a light bulb | Hacker News Reader

HD video demonstrated streaming over a light bulb

geek.com

30 pointsukdm22 comments

This demo looks at a first glance convincing, but it is a one way data stream. The point of using existing infrastructure to fill the room with high speed data reminds me of all the R&D and attempts made by infrared LEDs. The line of sight (LoS) problem and interference by sunlight turned even simple applications like wireless speaker systems to complicated technologies. The supposedly simple ICs with an AGC (auto-gain-control) turned to complicated over 110dB solutions with all kind of stability problems. Extracting the datastream from multipoint sources without data corruption, the increased power demand for longer distance communication out of mobile devices were all application problems, which did not allow to get that solution of the ground. Dr. Haas lab solution looks great, but for me considering the idea form a practical implementation point of view using visible light via LEDs is a far distance.

One of the (many) issues with IrDa is that there are severe restrictions on the power output, since the radiation is invisible. Using visible light, especially in the context of general purpose lighting, allows you to achieve much higher SNR.

I suppose at night someone could pick this up if there is a window in the room being used, but this is no different than normal wifi which can be picked up outside the house.

More of the problem is that it doesn't work room to room, only line of sight like with a remote control. Could be useful sending to wireless speakers from the media player, if the player is in a different location than the TV for some reason, but in the same room. I suppose it breaks down if you add a second device unless each one grabs its own specific light frequency and you have narrow optical filters able to distinguish between them.

A simple CDMA or TDMA (code/time division multiple access) scheme allow multiple devices to receive/transmit at the same time. This is how cell phones work.

See: http://en.wikipedia.org/wiki/Code_division_multiple_access

>More of the problem is that it doesn't work room to room, only line of sight like with a remote control.

If you could cheaply make the lights transmit and receive, and if the data were recoverable at very low illumination, you could employ mesh routing -- there's often quite a bit of overlap from lightbulb to lightbulb.

But you're correct -- answering the LoS problem is significant. This could be very useful in many applications, but it's no silver bullet.

While the LoS aspect is potentially desirable in specific applications, I agree that initially it can be perceived as a limiting factor. Researchers at Boston University,Rensselaer Polytechnic Institute, and Nakagawa Lab in Japan are working with software tools to characterize non LoS applications using reflective and refractive surfaces found indoors. A powerful system can be envisioned if someone managed to utilize these principles with a device that contains a receiver with a wide view.

Or you have a "networked" light socket in each room and a discovery protocol to figure out what room to route the data to. More work on the initial infrastructure side, but then you don't have to worry about mesh- your room is covered.

You'd network your light sockets to get around the room-to-room issue. Each room would have a light bulb "port". Some simple communication could then figure out which room your in (similar to access-point hopping used in large wifi distributions)

I can see this as a potential technology for sideband data: stream the video over the light, but the control data would still go through WiFi (and then you wouldn't bog down your wifi with torrent downloads)

Yeah, I think so. So now you have to wire your house with a wired network, and when you want to watch TV or use your computer you have to turn the lights in all the rooms on.

This is not a new idea. Saw a functional demo more than eight years back. Checkout: http://www.talking-lights.com/

This seems to be different. In your link posted, the lights serve as simple positioning beacons. It seems analogous to the bandwidth we've come to expect from things like infrared remote controls.

In this case, and indeed the whole point of the talk, is they're transmitting HD video.

The concept of transmitting data through light has been around forever. For my undergrad physics lab class in college, my team used a laser to transmit audio. Another team did a similar thing except they hooked it up to computers and sent jpeg's. This was probably seven years ago, and this wasn't the first time that projects like this had been done for this class.

Lasers are completely different. Ever heard of fiber-optic communication? This is about using light for both illumination and communication at the same time.

Edison only had you beat by 100 years

https://secure.wikimedia.org/wikipedia/en/wiki/Photophone

At least since Alexander Graham Bell and the optical telephone: http://en.wikipedia.org/wiki/Photophone

Completely correct. This is not a new idea at all. However TalkingLights had a huge bandwidth limit because they used flourescent lighting.

LEDs can go to much higher speeds - but even still, this isn't new. Researchers in Japan and Boston University have been producing demos for years:

http://www.bu.edu/smartlighting/research/smart-light-prototy...

http://www.naka-lab.jp/index_e.html

This could also be useful for people who fear electro-smog.

Doubtable. Those people are not rational, they'll find a reason to not like it.

A high-speed pulsing current running nearby.... Should throw off radio and/or magnetic waves along with the light.

This could be the answer to the European's latest fear mongering scheme!

http://www.businessinsider.com/cell-phones-cause-cancer-2011...

Prof. Haas talked about using this technology in smartphones, but he didn't explain what would happen when my phone is in my (dark) pocket.

This would be useful for some ad-hoc broadcasts (whatever that would be), not sure about any other use.

Reply on news.ycombinator.com