Building a WiFi light stick

Forums › LAN (ethernet/Wifi/routers/Bluetooth) › Building a WiFi light stick

oscarp

4 posts

Wannabe Geek

Topic # 107631 15-Aug-2012 12:41
Hello, Im a design student looking to create a light installation for an event in Christchurch, the idea is to create a grid of light sticks with LEDs evenly distriuted that respond to the hidden WiFi networks. Ideally the light sticks would respond to the network signalls of peoples cell phones as they walk through the installation but im assuming it would be easier to use the local wifi networks. Something similar to the 'Wifi Light Painting' project, picture provided. Any ideas on how i would go about building the electronics for this installation would be greatly appreciated.

dontpanic42

1574 posts

Uber Geek
+1 received by user: 11

Reply # 672794 15-Aug-2012 13:07
Arduino might be able to do what you are after.

oscarp

4 posts

Wannabe Geek

Reply # 672807 15-Aug-2012 13:42
Arduino looks just like what im after. Would you have any idea what i would need as input sensor?

ubergeeknz

Fully Operational
3239 posts

Uber Geek
+1 received by user: 1044

Trusted

Vocus

Subscriber

Reply # 672808 15-Aug-2012 13:46
Here's my suggestion... 1. Get some wifi activated t-shirts (or just the controller box) 2. Hack 3. Profit! Or even: http://www.trademe.co.nz/computers/networking-modems/wireless-networking/other/auction-503302941.htm

joker97

16172 posts

Uber Geek
+1 received by user: 1863

Trusted

Reply # 672828 15-Aug-2012 14:13
Explain the hack and the profit bit?

chevrolux

2948 posts

Uber Geek
+1 received by user: 782

Subscriber

Reply # 672834 15-Aug-2012 14:17
Buy one of those Wifi finders and just use the LED pin outs on it to control relays or whatever to turn on your big light sticks

dontpanic42

1574 posts

Uber Geek
+1 received by user: 11

Reply # 672835 15-Aug-2012 14:18
Having the LEDs respond to WiFi signals from both nearby stationary APs, and potential WiFi signals coming from people's cellphones, does sound like the cheaper, and simpler, option. However, having them respond to the cell signals from people's phones would seem much cooler, and seem much more interactive. Perhaps you need to find out a bit more about the intervals at which cell phones emit (i.e. transmit) their own cell signals. Maybe someone on here could say whether using cell signals would be feasible.

ubergeeknz

Fully Operational
3239 posts

Uber Geek
+1 received by user: 1044

Trusted

Vocus

Subscriber

Reply # 672868 15-Aug-2012 15:00
Totally feasible. http://dx.com/s/RF+detector I'd say one of these set to the right sensitivity would work pretty well

oxnsox

1923 posts

Uber Geek
+1 received by user: 139

Reply # 672909 15-Aug-2012 15:52
+1 to chevrolux. Keep it simple and use the lowest cost detector unit you'll find. That way it'll be really broad band and get triggered by folks making calls/sending txt/tweets nearby. It'll be much more dynamic that way. Dormant when people are at a distance but active and dynamic the closer they get, and the more of them there are with active devices.

oscarp

4 posts

Wannabe Geek

Reply # 672975 15-Aug-2012 18:09
the detector is a great idea. have been exploring that. the one problem with the detecor is the number of exisiting leds it contains as i could only add extra leds to the existing frequency level. if i develop with an Arduino i can develop greater frequency range im assuming, the Arduino will obviously be far more complex but it will result in a more dynamic peice?

geocom

422 posts

Ultimate Geek
+1 received by user: 86

Subscriber

Reply # 673004 15-Aug-2012 19:29
You should be able to hack the detector to work with the Arduino. De-solder the LED's or just clip them off. Go out and buy some 3v Relays. Put the coil from the positive to the negative on the PCB where you just clipped off the LED. Then put the 5v on to one end of the relay switch and the other end into one of the digital pins. You can then use the Arduino to read when the led whenever the pin goes high. This also has the advantage that the Detecter and Arduino will be isolated from one another so the voltage of the detector will not matter. This has one disadvantage as you will be using up the digital pins to do this so the number of pins you can use for LED's will go down. So you may need some 8-Bit Shift Registers this will allow you to add at least 8 led's for only 3 pins on one package you can add more for the same number of pins. One other thing is that the output on the Arduino pins is only 40mA and LED's consume about 20mA each so if you want to use Rope Light or something like that you will need to use some MOSFETS this will allow you to switch on as many LED's as you need(Check the current rating of the MOSFET however as it will have a limit). Geoff Edit: If you use an addressable rope light, It will have 8Bit shift registers as that is how they work. Geoff E

geocom

422 posts

Ultimate Geek
+1 received by user: 86

Subscriber

Reply # 673056 15-Aug-2012 22:15
Thinking about it. Wi-Fi has an issue it has to be on and broadcasting. You will also get a ton of noise in some locations. A better idea for telling where a person is by something like IR LED's with IR Receiver's which can not be seen when the line is broken it tells the computer where the person is. Or you could use sound, or a camera with tracking software. Then pixel mapping the LED's around the subject. IR LED's will likely be the cheapest and easiest to do. But you may end up with bleeding from one set to another set. So you would need to make sure you set up the directional beam right or something to stop this from happening. Geoff Geoff E

oscarp

4 posts

Wannabe Geek

Reply # 673095 16-Aug-2012 01:17
Thanks everyone. Ill try and get my head around some of this stuff and see what i can produce. Geoff the about IR LEDs is interesting, it would work a lot better with the story i am try to communicate.