Polarisation is a property of electromagnetic waves, LEDs are unpolarised but you can stick a polariser in front of them which basically blocks all the light except for a particular polarisation. The polarisation is sort of the way up the light is, that is one way to look at it (though that will not help if you look at circular polarisation).
If you put a similar polariser in front of your sensor then it will only let the light through if it is suitably aligned. But to be honest I think this would be a bad way of doing things for inter-prop communications because it will cut the light you get out of the emitters greatly and alignment becomes an issue, polarisation is also affected by reflections etc.
As far as modulation goes, as long as the amplitude of the signals do not saturate the detector then the signal detected will be a sum of all the components aimed at them (30/50khz etc) then the unwanted stuff is filtered out.
But if the props are all sat next to each other why use IR at all?
Graham
p.s. What happened to that project with the cool box?
Polarization is a good idea that has its uses. For ten props used as a test bed for more, I think putting filters in front of IR transmitters/receivers will limit functionality, range, and raise cost and complexity (alignment, mounting, etc.). I would want some ease of turning it on and off in software and syncing it to other groups of IR and then blocking those same groups. My initial thought was a 30 and 50 mhz filter, then realized the 30 would need to selectively talk to the 50 at times and visa versa. The next step was to create the filter with software. One transmitter can serve software for 30 and 50 khz. But if 30 floods 50 and visa versa, I'm back to a consideration of pure IR controlled by software that determines who speaks and when, and who listens and when.
Yes, the props set next to each other so why use IR at all? Of course we could merely connect the serial interface wires and go at it. I did mention a test bed? One of the tests is for mobile emitters. IR is ideal for the mobile emitters that don't use wires. The chips can walk around and not tangle in wires because the communication is with light. They can also have some unusual mountings in some not-so-ordinary environments. In a ten prop test bed you cannot see the other reason. With some insane number of chips, you'll see it's certainly beneficial to get rid of so many wires. I am building the circuits with light, not wires.
The idea of using Near Field Magnetic Induction technology has been elevated to the "awesome idea" status (thanks jazz!) - and is still in the running but I need help designing a circuit that can transmit and receive. From what I know, the benefits (with a low frequency carrier) are very low power consumed, wireless, good range for the computer hive, high security, good data transfer rates, completely mobile, not affected by stray infrared or light floods and reflections.
Graham Stabler: "What happened to that project with the cool box?"
It's right here, completed, with almost 30 programs running with 4 languages.
I'm writing up the paper for it with all the details. I'm testing it like a whirlwind!
The write-up is going a lot slower than I anticipated. There are build details, parts
lists, how to operate sections, modes and features, formula, theory, software and
programming, special techniques, a dictionary and resources/downloads,
information, and those intricate schematics, etc.
I typically have several projects going on at the same time and view each as a
step on a ladder going up to the next project. I hope I don't fall!!!
Comments
If you put a similar polariser in front of your sensor then it will only let the light through if it is suitably aligned. But to be honest I think this would be a bad way of doing things for inter-prop communications because it will cut the light you get out of the emitters greatly and alignment becomes an issue, polarisation is also affected by reflections etc.
As far as modulation goes, as long as the amplitude of the signals do not saturate the detector then the signal detected will be a sum of all the components aimed at them (30/50khz etc) then the unwanted stuff is filtered out.
But if the props are all sat next to each other why use IR at all?
Graham
p.s. What happened to that project with the cool box?
Yes, the props set next to each other so why use IR at all? Of course we could merely connect the serial interface wires and go at it. I did mention a test bed? One of the tests is for mobile emitters. IR is ideal for the mobile emitters that don't use wires. The chips can walk around and not tangle in wires because the communication is with light. They can also have some unusual mountings in some not-so-ordinary environments. In a ten prop test bed you cannot see the other reason. With some insane number of chips, you'll see it's certainly beneficial to get rid of so many wires. I am building the circuits with light, not wires.
The idea of using Near Field Magnetic Induction technology has been elevated to the "awesome idea" status (thanks jazz!) - and is still in the running but I need help designing a circuit that can transmit and receive. From what I know, the benefits (with a low frequency carrier) are very low power consumed, wireless, good range for the computer hive, high security, good data transfer rates, completely mobile, not affected by stray infrared or light floods and reflections.
Graham Stabler: "What happened to that project with the cool box?"
It's right here, completed, with almost 30 programs running with 4 languages.
I'm writing up the paper for it with all the details. I'm testing it like a whirlwind!
The write-up is going a lot slower than I anticipated. There are build details, parts
lists, how to operate sections, modes and features, formula, theory, software and
programming, special techniques, a dictionary and resources/downloads,
information, and those intricate schematics, etc.
I typically have several projects going on at the same time and view each as a
step on a ladder going up to the next project. I hope I don't fall!!!
humanoido