S1 packages with rtl-wmbus

[xaelsouth]

S1 packages with rtl-wmbus

[alalons]

tested ok with real meter! great work

data captured with new version
S1;1;1;2021-02-02 18:21:26.000;0;20;74117991;0x30446532917911742a087a60000000046d2d13a222025ac30001fd0c06326cffff0dff5f0c0008233000c10613080bfef9

Data captured with im871a dongle
Received telegram from: 74117991
manufacturer: (LSE) Landis Staefa electronic (0x3265)
type: Heat Cost Allocator (0x08)
ver: 0x2a
device: im871a[00101517]
rssi: -97 dBm
driver: unknown!
telegram=||30446532917911742A087A60000000046D2D13A222025AC30001FD0C06326CFFFF0DFF5F0C0008233000C10613080BFEF9|+353

[xaelsouth]

No, it's not implemented. I don't have ever seen a device with S1 mode implemented. Can you record a bitstream with devices inside?

[raritos]

looks like S1 mode is the same as C/T mode, but the problem is C/T mode is 100k where as S1 is 32k, so i guess your iir filter needs to be tuned for 32k sampling.
i tried to get 868.30Mhz S1 devices thru your code, but looks like iir filter is broke with 32k sampling, since i think you tuned iir for 100k sampling , as you stated on your code, right?
if i can come up with iir filter for 32k sampling , looks like it will work, but i just guess :-)
What can i do to change your iir filter to accomodate 32k sampling.
BTW, i will get S1 bitstream

[xaelsouth]

I suspect there is more of effort to invest into it. 32k sampling is just one thing; additonally you have

• an another sync-pattern;
• Manchester coding insead of NRZ. So you don't need the IIR filter for clock recovering.

If you have recorded a signal from your device, i can look into it.

[raritos]

Thanx xael,
---sync-patern is no problem for you :-) you already get C/T frames with
const uint32_t ACCESS_CODE = 0b0101010101010000111101u;
for T1 frame n × (01) 0001110110 10010110 n ≥ 279 , you can catch it.
-- Manchester is no problem for you, i am sure.

I tried to get packages but even if i can catch frame start, trailing bits are not correct, thats why i stopped modifying your code.
you state,if i am not mistaken, that your iir filter is tuned for 100k sampling, where S1 frame is 32k, so i end up with wrong bits.

my devices have 16day rule from the start of a month, so for a couple of days, i will not be able to get any signal. after 2. of next month , i will get many samples for you to look into.
Thanx again.

[xaelsouth]

You can easily recalculate bandpass filter coefficients I have used for clock recovering with Octave, see: https://github.com/xaelsouth/rtl-wmbus/blob/master/filter/bp_iir_cheb1_800kHz_90kHz_98Hz_102kHz_110kHz.m

Ws1 = 90e3/nyqistrate; % replace with (32768 - 10000)/nyqistrate
Wp1 = 98e3/nyqistrate; % replace with (32768 - 2000)/nyqistrate
Wp2 = 102e3/nyqistrate; % replace with (32768 + 2000)/nyqistrate
Ws2 = 110e3/nyqistrate; % replace with (32768 + 10000)/nyqistrate
Rp = 1; % leave it so
Rs = 40; % leave it so

Place new coefficients in the the function bp_iir_cheb1_800kHz_90kHz_98kHz_102kHz_110kHz(float sample).

Or better you create a new function named bp_iir_cheb1_800kHz_22kHz_30kHz_34kHz_42kHz(float sample) und use this instead of old one. Provide me with the patch if it works. But i would merge this after I have seen the bitstream and done some tests with that. Good luck! ;)

[alalons]

I have a test emitter from qundis capable of sending s1 data every 2 minutes. It's used to place amr collection devices and test signal strength.
If you give me the modified code I can try it anytime.

[xaelsouth]

Hello alalons,

could you record the signal for a period of about 2 or more minutes with

rtl_sdr s1_samples.bin -f 868.3M -s 1600000

and share s1_samples.bin to me?

[alalons]

Alvaro Alonso ha compartido un archivo de OneDrive con usted. Para verlo, haga clic en el vínculo siguiente. <https://1drv.ms/u/s!Am-8ChbmpLqisAO5QfiMdovxbJYK> [https://r1.res.office365.com/owa/prem/images/dc-generic_20.png]<https://1drv.ms/u/s!Am-8ChbmpLqisAO5QfiMdovxbJYK> s1_samples.tar.gz<https://1drv.ms/u/s!Am-8ChbmpLqisAO5QfiMdovxbJYK>

…

________________________________ De: Xael South <[email protected]> Enviado: lunes, 1 de febrero de 2021 19:39 Para: xaelsouth/rtl-wmbus <[email protected]> Cc: alalons <[email protected]>; Comment <[email protected]> Asunto: Re: [xaelsouth/rtl-wmbus] S1 packages with rtl-wmbus (#14) Hello alalons, could you record the signal for a period of about 2 or more minutes with rtl_sdr s1_samples.bin -f 868.3M -s 1600000 and share s1_samples.bin to me? — You are receiving this because you commented. Reply to this email directly, view it on GitHub<#14 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ASU3CPVYYYALFA7U66KP7CTS43YP7ANCNFSM4WXCFRXQ>.

[alalons]

those are the telegrams being transmitted alvaro@alvaro-Compaq-Presario-CQ61-Notebook-PC:~$ wmbusmeters --logtelegrams auto:s1 Started auto im871a[00101516] on /dev/ttyUSB0 listening on s1 Started auto rtlwmbus[00000001] listening on s1 No meters configured. Printing id:s of all telegrams heard! Received telegram from: 04998541 manufacturer: (LSE) Landis Staefa electronic (0x3265) type: Heat Cost Allocator (0x08) ver: 0x01 device: im871a[00101516] rssi: -37 dBm driver: unknown! telegram=||2C4465324185990401087A008000000B6E560400426C5F0C4B6E452301326C0000046D1A0E6F0202FDAC7E0123|+6 Received telegram from: 04998541 manufacturer: (LSE) Landis Staefa electronic (0x3265) type: Heat Cost Allocator (0x08) ver: 0x01 device: im871a[00101516] rssi: -37 dBm driver: unknown! telegram=||2C4465324185990401087A008000000B6E560400426C5F0C4B6E452301326C0000046D1A0E6F0202FDAC7E0123|+27 Received telegram from: 04998541 manufacturer: (LSE) Landis Staefa electronic (0x3265) type: Heat Cost Allocator (0x08) ver: 0x01 device: im871a[00101516] rssi: -37 dBm driver: unknown! telegram=||2C4465324185990401087A008000000B6E560400426C5F0C4B6E452301326C0000046D1A0E6F0202FDAC7E0123|+49 Received telegram from: 04998541 manufacturer: (LSE) Landis Staefa electronic (0x3265) type: Heat Cost Allocator (0x08) ver: 0x01 device: im871a[00101516] rssi: -37 dBm driver: unknown! telegram=||2C4465324185990401087A008000000B6E560400426C5F0C4B6E452301326C0000046D1A0E6F0202FDAC7E0123|+71

…

________________________________ De: Xael South <[email protected]> Enviado: lunes, 1 de febrero de 2021 19:39 Para: xaelsouth/rtl-wmbus <[email protected]> Cc: alalons <[email protected]>; Comment <[email protected]> Asunto: Re: [xaelsouth/rtl-wmbus] S1 packages with rtl-wmbus (#14) Hello alalons, could you record the signal for a period of about 2 or more minutes with rtl_sdr s1_samples.bin -f 868.3M -s 1600000 and share s1_samples.bin to me? — You are receiving this because you commented. Reply to this email directly, view it on GitHub<#14 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ASU3CPVYYYALFA7U66KP7CTS43YP7ANCNFSM4WXCFRXQ>.

[xaelsouth]

checkout and test

[xaelsouth]

Thanks for response! Then I will close the issue.