2
\$\begingroup\$

I have a use case where I want to place a small video camera element at a "remote" location and be able to record the video from it. The issue is that I can't run any wires to it, it can't transmit in any RF band and there is a very good chance whatever I use at that end will have to be disposable.

The solution I'm considering is to build a support circuit that gets the camera to dump the video to a serial line and use that to drive one or two LEDs (TxD and maybe clock). With the right optics and a pickup, I an use that as a simplex link. The question then is: how simple can I make the support circuit?

An Ideal solution would consist of a battery, an oscillator and the LED(s). I suspect I might need some amps, maybe a second clock and likely a little logic. What it would take likely depends a lot on how the interface to the selected camera works. (BTW, Googleing for SPI cameras and not getting Spy cameras is non-trivial.) I expect the camera and the battery will more or less consume the whole budget so the rest is going to have to be dirt simple and cheap.

Is this a viable project?


Edit: The kind of complexity, price and size (and video quality) I'm thinking of is something on the scale of this device. Note that I'm not including the dev board to plug it into as that would be one the receiving end of the link (and thus not potentially disposable). OTOH, from the images on that link, it seems the camera element has a ribbon cable attachment and I'm guessing is not a serial device.

\$\endgroup\$
  • \$\begingroup\$ What kind of frame rate and resolution do you want? Optical communication can support high data rates, but it's not just an op-amp, an LED, and a little logic, unless you've got really low requirements. You might want to start with something like the Leopardboard to learn more about digital imaging before trying to send it over an optical link... \$\endgroup\$ – Kevin Vermeer Jan 6 '12 at 21:31
  • \$\begingroup\$ @KevinVermeer: That would be step two. Step one would be finding out if what I want is even remotely feasible, thus this question. \$\endgroup\$ – BCS Jan 6 '12 at 22:45
2
\$\begingroup\$

If you want the ballpark value of a "good" free-space optical link, "Data rates of over 100 Mbit/s can be easily achieved" from the wikipedia page Free-space optical communication.

Using a visible light laser and receiver will probably work rather well. Though, to get any good data rates, you probably need to spend quite a lot of time on your design, or buy an already implemented laser receiving-optics + amplifier + demodulator module.

\$\endgroup\$
  • \$\begingroup\$ The constraints on the receiving end are much less restrictive so an off the shelf solution is ideal there. \$\endgroup\$ – BCS Jan 6 '12 at 23:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.