Boston University Working On LED Wireless Networks
Madas writes "This article on Absolute Gadget details how researchers at Boston University's College of Engineering are working on devloping wireless networks that use LED lights instead of normal radio waves. This research apparently has other uses in the automobile industry. Apparently the LEDs could warn you if the driver in front has put the brakes on so could avoid hitting the car in front. Personally, I'd use the vision balls that are in my thought box."
Erm, they should take a look at this:
http://ronja.twibright.com/
instead of re-researching it from scratch.
And the project is opensource.
Woyteck
An interesting idea, stupid article, and even worse Slashdot summary. For those who couldn't read all 5 paragraphs of the article, the idea is that LEDs can be rapidly modulated, basically acting like an IR remote, only in the visible spectrum. And they can modulate so fast that it's imperceptible to the eye (AKA "vision ball")
The brake light idea that the summary innaccurately mocks would actually allow the brakes in your car to be activated when your car "sees" the brake lights on the car in front of you activate. While this is a phenomenally stupid idea, it is different from what the summary indicates. I don't know why that bothers me, should be used to that by now.
The article also states that this technology would allow devices in your home (assuming they're equipped with LEDs) to wirelessly communicate directly with you, but doesn't say how. Morse code, perhaps?
Never let a lack of data get in the way of a good rant.
I knew I had seen an led-based point-to-point networking system described somewhere, and after a few minutes on hackaday, here it is, straight from 2005. Best part is, the linked to Ronja project is open, free speech-wise (and free beer for the major league scrounger).
Luke, help me take this mask off
I recall that 802.11 contains a part that specifies using infrared for 1 and 2Mbit/s multipoint operation. Apparently since its inception in 1997. Getting a bit more bandwidth out of it would be nice though. And, uh, more implementations.