disturbance of Infrared touch screen

[Evan Chow]

Hi,guys
I have been doing the project of Infrared touch screen for months,the touch signal is just good except the disturbance.The Infrared Emitting Diode and the Infrared Receiving Diode are ranked in two rows respectively.The distance is about 40.15 inch.Because the angle of Infrared Diodes is a little wide(25deg),adjacent Infrared Emitting Diode（the distance is about 0.43inch） will disturb each other(the Infrared Receiving Diode).
How to solve this problem?
Tks!

 

[jpanhalt]

Directionality can be improved by placing the diodes at the end of IR opaque tubes. I would tend to work on directionality of detection. In fact, I would consider putting the emitters close enough together to give a collimated bean across the area.

You do not mention the type of detector you are using. Are they digital or do they give an analog signal? If digital, you have the tube method to get directionality. You can also look at the signal to noise ratio for adjacent detectors. The one with the greatest S/N is probably nearest where the touch is. That method was used quite awhile ago in this robotics application: Robotics Projects Scroll to the 12C508 project. Also check out the DPRG link that is there.

If analog, it might be more effective to apply that same principle just looking at relative intensity.

John

 

[Evan Chow]

Thanks for jpanhalt!
It's analog signal.It seems that I haven't got too much time to change my hardware except the software.
I am trying to find some algorithm to save my signal.

 

[jpanhalt]

Are your emitters continuous or pulsed? If pulsed, can you describe their characteristics? Are you allowed to change the pulse pattern(s) for adjacent emitters?

John

 

[Evan Chow]

Thanks alot!
It's pulsed.The pulse width is about 300us and the duty is less than 1%.Only one pair diodes will work at a moment.