Need help decoding OOK Manchester

[Obl Ogis]

Hello,

I have captured this signal with rtl_433 which is said to be OOK Manchester.

Can someone help me building the flex decoder?

Thanks in advance!

[Christian Z.]

Doesn't look like Manchester. Also it's clipping (too "loud") which distorts the demodulation.

[Karl Lohner]

Using [rtl_433 -r g013_433.92M_250k.cu8 -R 0 -X "n=UNK,m=OOK_PCM,s=216,l=216,g=648,r=864,t=12"]  I see the raw bits from this sample, but due to signal quality these bits seem incorrect in spots.

The actual bit decoding should be something like "{360}aaaa9c56595556556665a66659556956a6595a9a6a999a6a595555565a9aaaa6aa555565a5559655595a5a65a4" which may be some kind of Manchester encoding.

https://triq.net/bitbench#c=%7B361%7Daaaa9e575d5557557775e7675d55e977a75dda9a7a9dde7a5d5555575fdfeacfad5555d795575d55656969d7940%20%5Bg013_433.92M_250k%20m%3DOOK_PCM%2Cs%3D216%2Cl%3D216%2Cg%3D648%2Cr%3D864%2Ct%3D12%20row%3A1%5D&c=%7B363%7Dbaaa9e5e5d555e557675a6667d55795fa75d7d3cf53bbcf4baaaaaaebd3d5d67aa555575e55597555d5e5e65a50%20%5Bg013_433.92M_250k%20m%3DOOK_PCM%2Cs%3D216%2Cl%3D216%2Cg%3D648%2Cr%3D864%2Ct%3D12%20row%3A2%5D&c=&c=%7B360%7Daaaa9c56595556556665a66659556956a6595a9a6a999a6a595555565a9aaaa6aa555565a5559655595a5a65a4%20%5Bmanual%20decode%20row%201%262%5D&f=hh%20&a=Preamble&m=38&d=DMC&cw=4

On Tue, Jan 18, 2022 at 11:13 AM Christian Z. <chri...@zuckschwerdt.org> wrote:

Doesn't look like Manchester. Also it's clipping (too "loud") which distorts the demodulation.

--
You received this message because you are subscribed to the Google Groups "rtl_433" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rtl_433+u...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/rtl_433/c050235a-289f-4db9-98d7-10f55ed25a18n%40googlegroups.com.

[Christian Z.]

Great job manually demodding the signal! Could really be DMC. I was fooled by what seemed length 3x and 4x pulses ;)

[Obl Ogis]

Thanks much Karl! There is the preamble I was looking for: 'aaaa9c', though the rest seems to be garbage.

Is there any rtl_433 parameter that I can use to obtain a better signal?

[Karl Lohner]

Your signal is very close to the center frequency you've tuned to, 433.92M, but this isn't actually desirable to avoid interference.

Try offsetting the center frequency by 100kHz or so one way or another, for example -f 433.8M or -f 434M.  You might also see if a sampling rate higher than 250k improves anything.