WiFi light painting | Hacker News Reader

WiFi light painting

nearfield.org

85 pointsKhalilK15 comments

I've always wondered if it's possible to build a video camera that films on the microwave spectrum instead of the visible light spectrum. Would such a tool would allow you to "see" wi-fi access points ? I guess they would look like bright, pulsating sources of light, and they would be visible through walls.

I did a bit of research on the subject, and they don't seem to really exist. It's possible that the longer wavelength poses technical difficulties. The individual receptors are too big and you can't cram them together at a sufficient density. Also, the lensing and shielding needs to be completely re-thought.

There's a 2012 paper called "Portable Real-Time Microwave Camera at 24 GHz" but i couldn't access it (not in academia).

Check out this article http://www.wired.com/2013/01/new-metamaterial-camera/

Also, try searching Synthetic Aperture Radar :) fun stuff

Here's the full text: http://cl.ly/1E3t000K2q0E

There are two ways to capture an EM signal. The static-array method, where you have something like a CCD (a matrix of antennas that sample a signal in parallel), is basically only useful around the visible-light part of the EM spectrum.

What you do for the rest of the spectrum is to take one antenna and move it around. As long as the signal is relatively static with respect to time, this has the same effect (and is much cheaper to implement.)

> The static-array method, where you have something like a CCD...is basically only useful around the visible-light part of the EM spectrum

Why is this the case? Is it just that technology is further along for visible light because there's more economic incentive for a digital camera that replicates the human eye?

Is it a materials problem, where we haven't discovered arrangements of matter with the right properties (e.g. CCD's respond to visible wavelengths and are adaptable to semiconductor manufacturing techniques)?

Or is it something to do with fundamental physics like the wavelengths are a lot longer which requires detectors that are too large to be practical? Or maybe diffraction is a problem?

Here's what an image from the camera described in that paper looks like: http://i.imgur.com/3z8Uae9.png

I wonder how much of the variation is temporal instead of spatial. There are a few shots in there that were done twice with different results, showing that the signal strength in a single spot varies over time. You can't use light painting to capture that usefully; you can only make pretty artistic impressions.

It looked like they were setting the device to respond to a specific ssid, so I assumed those shots were showing two different networks.

Also, you could use light painting to capture a time-varying signal: just do it a whole bunch of times in the same spot and turn them into individual frames in a video as in a time-lapse. It would just take a really long time. A motorized track for the sensor would make it a lot easier.

I think pretty artistic impression was the point. Project tried to reconcile art with technology that humans can't perceive. By appealing to our aesthetics this drives home the point that we are living surrounded by information fields that we can't perceive and give little thought to.

You make a good point, but I think you attached the wrong conclusion.

This is definitely time-relative. But you could use light-painting to capture that.

Just add a dimension; brightness, color, size. It just, obviously, requires more time. It won't be perfect, but that wasn't even remotely the point here.

This reminds me of a project, that a group of us talked about, but never did.

I used to work on a radio system with a very large and powerful phased array transmitter, the ground mat being hundreds of metres on each side. We though it would be fun to stick a bunch of fluorescent tubes into the ground. In theory we should have been able to see the antenna beam sweeping across the ground mat, as it was steered in difference directions and the tubes lit up. It would have been a very cool demonstration, but like so many other projects we only talked about it.

Great idea!

I'd love to see that in 3D... maybe with drones flying around instead of a single human doing the measurements?

I remember seeing this, though I still think that it wouldn't be so helpful, since I almost never have any luck with public wifi.

Same, much better off tethering to my phone here in the US. Maybe they could do it for carriers with cell signal and the carriers could improve any gaps, though. Same for private WiFi at a building or company.

Ever since seeing this in 2011 I wanted to use it to find signal shadow in the house. I have yet to build it...

Reply on news.ycombinator.com