Garbage NMEA output from GPS

[Subsentient]

So I've been fiddling with my phone, trying to get GPS working. From ModemManager, this is what I get after enabling NMEA.
Anyone else seeing this?
What exactly is required to get the GPS to actually work, since some people claim it works for them?

Used 'mmcli -m 0 --location-get' to produce output, after using --location-enable-gps-nmea.

[wibble]

I've had sensible looking output both from accessing the gps tty directly, and via gpsd, both under PmOS. I haven't checked with ModemManager in Mobian yet, beyond finding it needed a SIM inserted before it would accept commands.

[wibble]

Ok - sensible looking output in ModemManager too, but taking too long to get a fix. It took several minutes even to get sats listed as in view in GPGSV messages, and after 10 minutes there was still no fix. Previous experiments have been a quick check to see that I was getting NMEA out of the gps tty, not expecting a fix as I was doing it indoors. This time I took it outside, so sky view but still within wifi range, phone with screen to the sky and powered from a USB battery. This is with Mobian, last updated Friday, installed on eMMC, with a SIM but without uSD.

If past experience is anything to go by, first fix without priming with downloaded ephemeris data is the trickiest phase, and local noise sources can push the signal to noise ratio just the wrong side of being able to get a fix. As an example the Openmoko GTA02 needed the drive strength of the uSD connection reducing before reliable gps fix was possible.

---

Code:

  GPS  |               nmea: $GPGSA,A,1,,,,,,,,,,,,,,,,*32
       |                     $GPRMC,,V,,,,,,,,,,N*53
       |                     $GPGSV,3,1,09,08,57,286,23,20,41,067,26,21,45,068,25,27,82,111,27,1*6A
       |                     $GPGSV,3,2,09,01,02,237,,07,17,291,,10,48,116,,11,27,260,,1*68
       |                     $GPGSV,3,3,09,13,04,002,,1*59
       |                     $GPVTG,,T,,M,,N,,K,N*2C
       |                     $GPGGA,,,,,,0,,,,,,,,*66

Still no fix...or time for that matter

[wibble]

On the 3rd attempt leaving the phone outside in between rain showers I finally got a fix. This shows it can work without assistance, but it's not exactly quick or reliable. The assistance process is detailed in EC25&EC21_GNSS_AT_Commands_Manual_V1.1 - see sections 2.7 through 2.9, a d examples in 3.3. The file command used to upload the assistance data is in EC2x&EG25-G&EG9x& EM05_FILE_AT_Commands_Manual_V1.0 section 2.4. If assistance isn't already enabled it requires a modem restart too.

[Dmytro]

Code:

Trying to get fix...

Still no fix...or time for that matter

Time is definitely very important factor here. For low frequency devices fix can take hours without assistance especially after long inactive move (aka shipment from China).

[wibble]

Indeed - I just haven't dug out an old GPS to use as a reference with the same sky view yet. It's not an unobstructed view, so I could just have been unlucky with the sat positions earlier.

[Dmytro]

Has anybody succeeded to gen nmea data by gpsd from modem?
For me --location-get works (more or less) but gpsd has nothing. I have it attached to ttyUSB1. Is it right?
Others available are:
USB: ttyUSB0 ttyUSB1 ttyUSB2 ttyUSB3
Serial: ttyS0 ttyS2 ttyS3

[wibble]

Has anybody succeeded to gen nmea data by gpsd from modem?
For me --location-get works (more or less) but gpsd has nothing. I have it attached to ttyUSB1. Is it right?
Others available are:
USB: ttyUSB0  ttyUSB1  ttyUSB2  ttyUSB3
Serial: ttyS0  ttyS2  ttyS3

ttyUSB1 is correct for nmea output, and ttyUSB2 for enabling gps with AT+QGPS=1 and any configuration you care to do. I haven't tried gpsd on mobian, but I think it worked in PmOS. You can check it with picocom or similar.

[Dmytro]

ttyUSB1 is correct for nmea output, and ttyUSB2 for enabling gps with AT+QGPS=1 and any configuration you care to do. I haven't tried gpsd on mobian, but I think it worked in PmOS. You can check it with picocom or similar.

I was able to propagate nmea data to application layer.
But, receiver is so weak that it get fix only on wide empty space. It seems even clouds are not welcome. 
And it takes time regardless assisted or not. I'll experiment tomorrow trying to refix on the same position.
I'm afraid it may be antenna issue or even worth.
Anyone has different experience?

[Subsentient]

Sadly I think you may be right. It's nearly impossible for me to get a fix, but even when I do, the ModemManager output remains garbage. I'm able to see data from gpsd on the rare occasion it does get a fix, however.