Which IR protocol is this?

[DaGlitch]

Been trying to figure this one out. I'm looking at a simple IR transponder unit. The image below is the raw IR coming from it. It had the number "3420" written on it, so I know that will correlate to somewhere in the data. My guess is that it's 20 bits possibly? Carrier frequency looks to be ~ 460kHz. The numbers I have written above each chunk is the pulse count which might be +/- 1. Then time in between each pulse. The 2nd image is how I imagine the way each bit is broken up (which may be incorrect).

[bson]

Looks bi-phase encoded.  RC-5 perhaps?

3420 is hex d5c, or binary:
   1101 0101 1100

Your pulse train (mark = 0, space = 1) divided into ~20usec clock periods

0,1,0,1,1,0,0,0,0,1,0,0,0,1,1,1,0,1,0,1,0,1,0,0,0,0,0,1,1,0,1,0,1,0,0,1,1,0

Bi-phase decoded (transition during clock = 1, no transition = 0):

   1    1    1   0    1    0    1    0    1    1    1   0    0    1    1    1    1    1   1
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^

There you go.

http://ckp.made-it.com/encodingschemes.html

[DaGlitch]

Thanks for the response,

I guess the only difference I have in what you have for the pulse train is the 1 I have indicated:
0,1,0,1,1,0,0,0,0,1,0,0,0,1,1,1,0,1,0,1,0,1,0,0,0,1,0,0,1,1,0,1,0,1,0,0,1,1,0,1
                                                                                 ^

But maybe I'm just not getting the decoding part as I came up with:
1  1  1  0  1  0  1  0  1  1  1  0  1  0  0  1  1  0  0  1
    ^^^^^    ^^^^^   ^^^^^
        D               5             D?

[DaGlitch]

Any other ideas?

[Dave Turner]

Have a look at the IR decoders available for the arduino. There's some useful info there that might help.

[JoeO]

It might help if you supplied more information.  What is it?  Who is the manuf.?  Picture.

[bson]

To decode bi-phase, make sure you sample the levels correctly at the recovered clock.

[jimmc]

Dug this out of my bookmarks, might be of interest (I haven't tried it myself) ... http://wiki.nethome.nu/doku.php/analyzer/start

Jim

[DaGlitch]

It's essentially this. it's an ir transponder.

It might help if you supplied more information.  What is it?  Who is the manuf.?  Picture.

[macboy]

To me, it looks like it could be UART (async serial) data at 38400 using OOK (on-off keying) with "on" (carrier present) being Low or zero and "off" (no carrier) being a High or one. With TTL level UART, idle is High state and the start bit is a Low, and stop bit is a High. Data bits are non-inverted (Low=0).

I say that it might be serial data because there is a start bit (on) followed by 8 various bits, followed by a stop bit (off), and this is repeated 4 times (40 bits total with 32 data bits), without framing errors, i.e. the start and stop bits always fall in the right place. And a ~26 us bit period is very close to 38400 baud, which is a very common baud rate.

Assuming carrier present is "0" and idle is "1", then the bit sequence might be:
0101100001000111010101000100110101001101(11111....)

If you group the bits into UART-style framing with a "0" start bit, eight data bits, and a "1" stop bit, then you see that you have:
0 10110000 1
0 00111010 1
0 10001001 1
0 10100110 1
... followed by idle, a valid state.

UART data is sent LSb first, so reverse the bits in each byte and the data becomes:
b00001101 = 0x0D
b01011100 = 0x5C
b10010001 = 0x91
b01100101 = 0x65

Notice that 0x0D5C is decimal 3420. It seems to me this is probably your answer!

(double check bits sequence and hex conversions)

p.s. 455 kHz is a common IR modulation frequency, and you can buy nice little receiver/demodulators for that frequency. If you plan to receive the data from the transponder, then use one. It will be far less sensitive to noise and more reliable than anything you could make yourself.