Decoding new door/window sensor

[rbhcsd]

I have rtl-433 installed on a PiZero feeding outdoor temperature sensor data to my Home Assistant Pi. I recently purchased some cheap, no logo, 433MHz magnetic sensors (door/window) to use but they are not decoding automatically. I am not a SDR user outside of this involvement, so not sure how to go about building a decoder.

I installed rtl_433 on my MacBook using HomeBrew with an additional SDR dongle.

I ran the "rtl_433 -f 433.92M -S unknown" command with the antenna removed and the sensor located next to the SDR.

----OUTPUT from command-------------------
rtl_433 version 24.10 (2024-10-30) inputs file rtl_tcp RTL-SDR with TLS
Found Rafael Micro R820T tuner
âSDRê Using device 0: Realtek, RTL2838UHIDIR, SN: 00000001, "Generic RTL2832U OEM"
Exact sample rate is: 250000.000414 Hz
âR82XXê PLL not locked!
*** Saving signal to file g004_433.92M_250k.cu8 (417275 samples, 917504 bytes)
îCSignal caught, exiting!

I then ran the "rtl_433 -A g003_433.92M_250k.cu8" against the output file, reviewed the output, and went to the link provided in the output below.

----OUTPUT from command-------------------
rtl_433 version 24.10 (2024-10-30) inputs file rtl_tcp RTL-SDR with TLS
âInputê Test mode active. Reading samples from file: g004_433.92M_250k.cu8
Detected OOK package à0.231444s
Analyzing pulses...
Total count: 24, width: 35.80 ms ( 8951 S)
Pulse width distribution:
â 0ê count: 1, width: 148 us â148;148ê ( 37 S)
â 1ê count: 12, width: 444 us â440;452ê ( 111 S)
â 2ê count: 11, width: 1164 us â1156;1176ê ( 291 S)
Gap width distribution:
â 0ê count: 12, width: 412 us â388;420ê ( 103 S)
â 1ê count: 11, width: 1136 us â1128;1144ê ( 284 S)
Pulse period distribution:
â 0ê count: 1, width: 536 us â536;536ê ( 134 S)
â 1ê count: 22, width: 1580 us â1572;1592ê ( 395 S)
Pulse timing distribution:
â 0ê count: 1, width: 148 us â148;148ê ( 37 S)
â 1ê count: 24, width: 428 us â388;452ê ( 107 S)
â 2ê count: 22, width: 1148 us â1128;1176ê ( 287 S)
â 3ê count: 1, width: 11764 us â11764;11764ê (2941 S)
Level estimates âhigh, lowê: 15955, 25
RSSI: -0.1 dB SNR: 28.0 dB Noise: -28.2 dB
Frequency offsets âF1, F2ê: 1322, 0 (+5.0 kHz, +0.0 kHz)
Guessing modulation: Pulse Width Modulation with sync/delimiter
view at https://triq.org/pdv/#AAB104009401AC047C2DF48192A1A1A19292A1A1929292A1A192A19292A192A1A1929355
Attempting demodulation... short_width: 444, long_width: 1164, reset_limit: 1148, sync_width: 148
Use a flex decoder with -X 'n=name,m=OOK_PWM,s=444,l=1164,r=1148,g=0,t=0,y=148'
âpulse_slicer_pwmê Analyzer Device
codes : é23è8ce5a6

Detected OOK package à0.280220s
Analyzing pulses...
Total count: 25, width: 38.45 ms ( 9612 S)
Pulse width distribution:
â 0ê count: 13, width: 448 us â440;460ê ( 112 S)
â 1ê count: 12, width: 1164 us â1160;1176ê ( 291 S)
Gap width distribution:
â 0ê count: 12, width: 1132 us â1128;1140ê ( 283 S)
â 1ê count: 12, width: 416 us â412;420ê ( 104 S)
Pulse period distribution:
â 0ê count: 24, width: 1580 us â1572;1600ê ( 395 S)
Pulse timing distribution:
â 0ê count: 25, width: 432 us â412;460ê ( 108 S)
â 1ê count: 24, width: 1148 us â1128;1176ê ( 287 S)
â 2ê count: 1, width: 11764 us â11764;11764ê (2941 S)
Level estimates âhigh, lowê: 15933, 9
RSSI: -0.1 dB SNR: 32.5 dB Noise: -32.6 dB
Frequency offsets âF1, F2ê: 1022, 0 (+3.9 kHz, +0.0 kHz)
Guessing modulation: Pulse Width Modulation with fixed period
view at https://triq.org/pdv/#AAB10301B0047C2DF48190819090908181909081818190908190818190819090818255
Attempting demodulation... short_width: 448, long_width: 1164, reset_limit: 1144, sync_width: 0
Use a flex decoder with -X 'n=name,m=OOK_PWM,s=448,l=1164,r=1144,g=0,t=286,y=0'
âpulse_slicer_pwmê Analyzer Device
codes : é25èa339698
-------------(this repeated several times, this is only the first occurrence)-------------

I visited the website via the link and noted the output from two different sensors noting the "Bits {25}" entry was different, so I assume that value is unique to each sensor.

At this point, I don't know how to build a decoder and have it included in the rtl_433 decoder listing so I can turn these messages into MQTT json messages to HA.

Can someone tell me how to proceed? Sorry I'm so new to SDR I'm not sure what to do next and I'm only a novice linux user. Thanks for any assistance.

[zuckschwerdt]

That's a fixed bit-width PWM signal, the suggestion is okay. Use a flex decoder of
-X 'n=name,m=OOK_PWM,s=430,l=1250,r=1500,bits>=20'
to grab codes.

If that works well then look through the conf folder and write a flex decoder conf like one of those.
If you only want to capture those two sensors then two decoders, each with e.g. a match={25}a339698 will work.

[AkaBoing]

code never changes..
good for you, bad for us, this dont help us identify a mb. ID Position in the data.
but what happens when battery gets empty? lol, a CR2032 can hold for 10 years and more, so we prob. get no answer to this.
i play with the data, invert, reflect and so on to get a CHS ,CRC or whatever but no result for now.
stay tuned..
the white point on the front cover (the rounded one) , whats this ? , a simple design point ?

edit: i think about the H1 LED, if this is a infrared Receive Diode, then ..
a simple check with a RC5 Code remote control..
but i have no hope..

[rbhcsd]

The white circle on the front allows you to push it in and set off the unit (transmit) without moving or having the magnet next to it. It's probably something to do with getting the unit connected to some type of alarm base unit. I verified that pushing the button gives you the same data as removing the magnet. Attached is what I captured by just pressing the button without the magnet around.
ButtonPushNoMagnet.txt

[AkaBoing]

thx , same data, so not really helpfull.
a 24bit puzzle..

[rbhcsd]

Just thought I would do a final update. I received the other magnet sensors today in the mail and they do show a different code for open and close as well as each sensor has a different id. They come up as Generic-Remote. I've attached the output of two different sensors with an open and close event each. These will will work for my project, just have to test the distance factor from my rtl-433 pizero. Thanks for all the help and I've learned quite a bit about rtl-433!!
PS - like the other sensor they send each message at least 30 times.
NewSensors.txt

[AkaBoing]

good work..
looks like a EV1527 variant.
dont close this threat, i want to knew the final..