[Ovmsdev] v2 simcom, gsm, and ancient protocols
Arthur Hebert
ahebert at gmail.com
Wed Oct 11 23:40:44 HKT 2017
Awesome! Thanks for posting about it, Mark.
-Arthur
On Tue, Oct 10, 2017 at 8:15 PM, Mark Webb-Johnson <mark at webb-johnson.net>
wrote:
> Working 95% now:
>
> I (25695) gsm-mux: Start MUX
> I (25695) gsm-mux: StartChannel(0)
> SIMCOM tx f9 03 3f 01 1c f9 ..?...
> SIMCOM rx f9 03 73 01 d7 f9 ..s...
> I (25705) gsm-mux: ProcessFrame(CHAN=0, ADDR=03, CTRL=73, FCS=d7, LEN=6)
> I (25705) gsm-mux: Channel #0 is open
> I (25705) gsm-mux: StartChannel(1)
> SIMCOM tx f9 07 3f 01 de f9 ..?...
> SIMCOM rx f9 07 73 01 15 f9 ..s...
> I (25705) gsm-mux: ProcessFrame(CHAN=1, ADDR=07, CTRL=73, FCS=15, LEN=6)
> I (25705) gsm-mux: Channel #1 is open
> I (25705) gsm-mux: StartChannel(2)
> SIMCOM tx f9 0b 3f 01 59 f9 ..?.Y.
> SIMCOM rx f9 0b 73 01 92 f9 ..s...
> I (25715) gsm-mux: ProcessFrame(CHAN=2, ADDR=0b, CTRL=73, FCS=92, LEN=6)
> I (25715) gsm-mux: Channel #2 is open
> I (25715) gsm-mux: StartChannel(3)
> SIMCOM tx f9 0f 3f 01 9b f9 ..?...
> SIMCOM rx f9 0f 73 01 50 f9 ..s.P.
> I (25715) gsm-mux: ProcessFrame(CHAN=3, ADDR=0f, CTRL=73, FCS=50, LEN=6)
> I (25715) gsm-mux: Channel #3 is open
>
> OVMS > simcom muxtx 3 AT
> SIMCOM mux tx 41 54 AT
> SIMCOM tx f9 0d ff 05 41 54 01 f9 ....AT..
> SIMCOM mux tx 0d 0a ..
> SIMCOM tx f9 0d ff 05 0d 0a 01 f9 ........
>
> SIMCOM rx f9 0d ff 0d 0d 0a 4f 4b 0d 0a 0f f9 ......OK....
> I (32015) gsm-mux: ProcessFrame(CHAN=3, ADDR=0d, CTRL=ff, FCS=0f, LEN=12)
>
> SIMCOM rx f9 0d ff 1f 0d 0a 2b 43 53 51 3a 20 31 35 2c 39 ......+CSQ: 15,9
> SIMCOM rx 39 0d 0a f0 f9 9....
> I (84005) gsm-mux: ProcessFrame(CHAN=3, ADDR=0d, CTRL=ff, FCS=f0, LEN=21)
>
> OVMS > simcom muxtx 3 AT+COPS?
> SIMCOM mux tx 41 54 2b 43 4f 50 53 3f AT+COPS?
> SIMCOM tx f9 0d ff 11 41 54 2b 43 4f 50 53 3f 1a f9 ....AT+COPS?..
> SIMCOM mux tx 0d 0a ..
> SIMCOM tx f9 0d ff 05 0d 0a 01 f9 ........
>
> SIMCOM rx f9 0d ff 4d 0d 0a 2b 43 4f 50 53 3a 20 30 2c 30 ...M..+COPS: 0,0
> SIMCOM rx 2c 22 20 43 53 4c 20 20 4a 73 79 20 54 65 6c 22 ," CSL Jsy Tel"
> SIMCOM rx 2c 32 0d 0a 0d 0a 4f 4b 0d 0a 7f f9 ,2....OK....
> I (125425) gsm-mux: ProcessFrame(CHAN=3, ADDR=0d, CTRL=ff, FCS=7f, LEN=44)
>
> OVMS > metrics list
> m.freeram 69196
> m.hardware OVMS WIFI BLE BT cores=2 rev=ESP32/1
> m.monotonic 134
> m.net.mdm.iccid 8944501003151XXXXXX
> m.net.mdm.model 3531XXXXSIM5360E
> m.net.provider CSL Jsy Tel
> m.net.sq -77
>
>
> Mark
>
> On 11 Oct 2017, at 8:57 AM, Mark Webb-Johnson <mark at webb-johnson.net>
> wrote:
>
>
> I’m having a fascinating time with the new SIMCOM modem support for OVMS
> v3.
>
> I’ve wanted to use the CMUX protocol for some time, as it allows the
> single async line to be split to multiple separate logic communication
> channels, and for each of those channels to be used for different things.
> So, we can:
>
>
> - Create a channel for the GPS. Then, that can be straight NMEA
> protocol, streaming from the GPS, with no need to poll.
> - Create a channel for the PPP connection used to provide data service
> (GPRS, etc). This means we can handle our own networking (with lwip) and
> expose the modem network as a pure network accessible to normal networking
> stacks in the ESP32 framework. A “connect” would work the same way, no
> matter if it is over Wifi, or GSM modem.
> - Create a channel for AT commands.
>
>
> The real advantage is that these are all separate, so we can issue AT
> commands while reading a GPS NMEA message, while sending telemetry to the
> server.
>
> Those of you familiar with OVMS v2 net.c code will recognise the issue;
> synchronising all the different accesses to a single async link used for
> AT, GPRS, SMS, and DATA, was a nightmare. CMUX was actually in the 2G
> SIMCOM chips as well, but we just didn’t have the ram or flash space in v2
> to do it.
>
> So, now v3 is here, and I start to work on it. Turns out that the CMUX
> protocol is not SIMCOM’s. Instead, they use an industry standard protocol
> called "GSM 07.10 Multiplexing protocol”, in basic mode. The specification
> I am using dates back to 1997. Also of note is that the AT command set has
> also been standardised by the same body as GMS 07.07. Reading that
> specification, and working with that protocol, is like a journey back in
> history. Most of it seems based on the X.25 protocol from back in the
> 1980’s (my I.T. career started with messing around with X.25 systems, and
> asynchronous RPADs, so it is dredging up old memories dealing with Set
> Asynchronous Balanced Mode (SABM) and Unnumbered Acknowledgement (UA)
> responses). Great fun…
>
> Anyway, it is a big specification, with lots of nuances, but I am making
> progress.
>
> To leave normal single channel AT command mode, and enter GSM 07.10
> multiplexor mode, we send some AT commands to program the multiplexor
> channel connections that we want:
>
>
> - Channel 0 is mapped to the DIAG service
> - Channel 1 is mapped to the NMEA (GPS) service
> - Channel 2 is mapped to the DATA (AT) service
> - Channel 3 is mapped to the DATA (AT) service
>
>
> Then we send “AT+CMUX=0” and wait for the OK:
>
> SIMCOM tx 41 54 2b 43 4d 55 58 53 52 56 50 4f 52 54 3d 30 AT+CMUXSRVPORT=0
> SIMCOM tx 2c 30 3b 2b 43 4d 55 58 53 52 56 50 4f 52 54 3d ,0;+CMUXSRVPORT=
> SIMCOM tx 31 2c 35 3b 2b 43 4d 55 58 53 52 56 50 4f 52 54 1,5;+CMUXSRVPORT
> SIMCOM tx 3d 32 2c 31 3b 2b 43 4d 55 58 53 52 56 50 4f 52 =2,1;+CMUXSRVPOR
> SIMCOM tx 54 3d 33 2c 31 3b 2b 43 4d 55 58 3d 30 0d 0a T=3,1;+CMUX=0..
> SIMCOM rx 0d 0a 4f 4b 0d 0a ..OK..
>
>
> Now, we start talking GSM 07.10. We need three channels, so send SABM
> messages and wait for UA acknowledgements, for each in sequence:
>
> I (88695) gsm-mux: Start MUX
> I (88695) gsm-mux: StartChannel(0)
> SIMCOM tx f9 03 3f 01 1c f9 ..?...
> SIMCOM rx f9 03 73 01 d7 f9 ..s...
> I (88705) gsm-mux: ProcessFrame(CHAN=0, ADDR=03, CTRL=73, FCS=d7, LEN=6)
> I (88705) gsm-mux: Channel #0 is open
> I (88705) gsm-mux: StartChannel(1)
> SIMCOM tx f9 07 3f 01 de f9 ..?...
> SIMCOM rx f9 07 73 01 15 f9 ..s...
> I (88705) gsm-mux: ProcessFrame(CHAN=1, ADDR=07, CTRL=73, FCS=15, LEN=6)
> I (88705) gsm-mux: Channel #1 is open
> I (88705) gsm-mux: StartChannel(2)
> SIMCOM tx f9 0b 3f 01 59 f9 ..?.Y.
> SIMCOM rx f9 0b 73 01 92 f9 ..s...
> I (88715) gsm-mux: ProcessFrame(CHAN=2, ADDR=0b, CTRL=73, FCS=92, LEN=6)
> I (88715) gsm-mux: Channel #2 is open
> I (88715) gsm-mux: StartChannel(3)
> SIMCOM tx f9 0f 3f 01 9b f9 ..?...
> SIMCOM rx f9 0f 73 01 50 f9 ..s.P.
> I (88715) gsm-mux: ProcessFrame(CHAN=3, ADDR=0f, CTRL=73, FCS=50, LEN=6)
> I (88715) gsm-mux: Channel #3 is open
>
>
> Once the channels are open, then the SIMCOM modem sends messages:
>
> SIMCOM rx f9 0d ff 1f 0d 0a 2b 43 53 51 3a 20 31 34 2c 39 ......+CSQ: 14,9
> SIMCOM rx 39 0d 0a f0 f9 9....
> I (403335) gsm-mux: ProcessFrame(CHAN=3, ADDR=0d, CTRL=ff, FCS=f0, LEN=21)
>
>
> That is a signal quality report (+CSQ) coming in on the data channel.
>
> Here is what an incoming SMS message (+CMT) looks like:
>
> SIMCOM rx f9 0d ff c3 0d 0a 2b 43 4d 54 3a 20 22 2b 34 34 ......+CMT: "+44
> SIMCOM rx 37 39 33 37 34 30 xx xx xx xx 22 2c 22 22 2c 22 793740xxxx","","
> SIMCOM rx 31 37 2f 31 30 2f 31 31 2c 30 31 3a 35 33 3a 35 17/10/11,01:53:5
> SIMCOM rx 31 2b 30 34 22 2c 31 34 35 2c 34 2c 30 2c 30 2c 1+04",145,4,0,0,
> SIMCOM rx 22 2b 34 34 37 37 39 37 37 30 xx xx xx xx 22 2c "+44779770xxxx",
> SIMCOM rx 31 34 35 2c 31 31 0d 0a 48 45 4c 4c 4f 20 57 4f 145,11..HELLO WO
> SIMCOM rx 52 4c 44 0d 0a xx f9 RLD..u.
> I (1727775) gsm-mux: ProcessFrame(CHAN=3, ADDR=0d, CTRL=ff, FCS=xx,
> LEN=103)
>
>
> It is a lot of work getting this working, but the result should be worth
> it.
>
> Regards, Mark
>
> _______________________________________________
> OvmsDev mailing list
> OvmsDev at lists.teslaclub.hk
> http://lists.teslaclub.hk/mailman/listinfo/ovmsdev
>
>
>
> _______________________________________________
> OvmsDev mailing list
> OvmsDev at lists.teslaclub.hk
> http://lists.teslaclub.hk/mailman/listinfo/ovmsdev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.teslaclub.hk/pipermail/ovmsdev/attachments/20171011/1b496672/attachment-0001.html>
More information about the OvmsDev
mailing list