Unknown TPMS

[edwin-nel]

Please help me decode a TPMS sensor?
I struggle to determine proper flex parameters, and also not sure how to figure out whether it is OOK or FSK.

All files is for one sensor, (in the Monitor menu, I can see the following identifier 85 E2 for the sensor )
Attached cu8 files, captured with a RTL-SDR dongle, 443MHz, 250kHz

pdv - g001 - 0.3 bar 31 celsius

pdv - g002 - 0.5 bar 30 celcius

pdv - g019 - 0.0 bar 31 celcius

Using -X, I've extracted the codes from multiple files with different pressure and temperatures.

The three cu8 files for the pdv analysis. g001, g002, g019 ... cu8

[zuckschwerdt]

The transmission seems to be OOK. There is a warmup and then preamble. The timing is 100 µs PCM, likely MC or DMC coded.
rtl_433 -R 0 -X 'n=name,m=OOK_MC_ZEROBIT,s=100,l=100,r=250' should work to get codes.
Or rtl_433 -R 0 -X 'n=name,m=OOK_PCM,s=100,l=100,r=250' to get raw data.

Try to get a range of codes for a slow ramp of pressure, that way we can find the value field.

[zuckschwerdt]

The data seems too random. Either the transmission is encrypted or it's DMC and not MC. Can you capture codes with OOK_PCM so we can try DMC in BitBench?

[zuckschwerdt]

Actually we can reverse the MC and get a BitBench to play with codings. We can't decide wether it's MC or DMC and the data is too random.

[edwin-nel]

When I start capturing with OOK_PCM, when the pressure is updated, the first signal is completely different than the rest of the following signals. Is that correct behaviour, should the group of 7 signals not be more or less the same?
28 Celsius 0.4 Bar

[edwin-nel]

Here are 3 updates from the sensor. bitbench

The first line for each burst is different than the rest. This is the same pattern for each update. You think the transmission can be encrypted?

[zuckschwerdt]

Those codes are all the same contents, it's just a longer preamble.
Try "Align" "5559" and "Sync", then "Manchester".