No fix location

[ferorted]

Dear Mictronics, I did many tests with your pluto-gps-sim software but it can't fix the location.

I'm using pluto-gps-sim in a Ubuntu PC, a Pluto SDR and a uBlox M8-T GPS receiver.

The uBlox receives the simulated signals but it can't fix the location.

I know that a strong signal received can be a problem, so i tried from -80 to -150dBm at the uBlox receiver input ( 70dB attenuator plus the pluto-sdr-sim Att.) I also tried with bandwith from 2 to 5MHz, but the results are always the same: No fix location.

Attached you can find an analyzer and ublox-center images. As you can see in the images, it is not much time with 12 satellites in green, then they go to blue. And this situation repeats all over again.

Could you help me? I will be very grateful with any idea.

[Mictronics]

Can you provide the command line you use to start pluto-gps-sim.

[ferorted]

Hello,

Thank you so much for your answer.

The command is:

./pluto-gps-sim –e hour0430.20n –l 40.4574, -3.6700,700 –T now -N 192.168.2.1 –v -s 2600000 –A –xx.x –B y.y

xxx.x is the attenuation, I tried some values

y.y is the bandwidth, also I tried from 2.0 to 5.0

Thank you so much.

---

De: Mictronics [mailto:notifications@github.com] Enviado el: jueves, 13 de febrero de 2020 14:34 Para: Mictronics/pluto-gps-sim CC: ferorted; Author Asunto: Re: [Mictronics/pluto-gps-sim] No fix location (#6)

Can you provide the command line you use to start pluto-gps-sim.

— You are receiving this because you authored the thread. Reply to this email directly, view https://github.com/Mictronics/pluto-gps-sim/issues/6?email_source=notificat ions&email_token=AOBCA55J3PDQLFQER6GLICDRCVD27A5CNFSM4KT67DI2YY3PNVWWK3TUL52 HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOELU6ZNQ#issuecomment-585755830 it on GitHub, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOBCA53N6XGJR3KXS23KH63RC VD27ANCNFSM4KT67DIQ . https://github.com/notifications/beacon/AOBCA52RDSNAXLX2GWNG5GDRCVD27A5CNFS M4KT67DI2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOELU6ZNQ .gif

[ferorted]

Hello again, Attached there are 2 images from a test with Measurement Period setting every 1000 msg. and 100 msg. With 100 msg. you can see that there are “a lot of gaps” it the received data of each satellite.

Can it be the problem? Do you know why it is and how fix it?

Thank you very much.

[Mictronics]

Please test with changes in branch "testing". The amplitude of the sine tables are now reduced to avoid clipping and distortion of the final IQ stream.

That's with a Navilock NL-800U2.

[ferorted]

Thank you so much for your support.

However, my tests are unsuccessful. U-center shows similar results than previous, even worse.

Measurements in spectrum analyzer show that the new signal generated by "testing" code are about 50dB below signal generated by "master" code. Also the "testing signal" is not constant. Please, see the attached videos.

Thank you and best regards,

signals.zip

[Mictronics]

What is your hardware setup? (in addition to pluto) With test code I get the same GPS satellite constellation indicated on a Samsung Galaxy Tab S2 with satstats app. However, no position lock because the tabled is receiving Glonass in addition. But GPS PRN's correspond to what is simulated.

The only thing I can imagine for an interrupted indication is that you IQ stream from host system to pluto is interrupted too, hence the buffer is running empty.

[ferorted]

My hardware setup is a Lenovo ThinkPad i3 laptop running Ubuntu 18.04.3 connected to USB port of PlutoSDR (v 0.31 firmware). PlutoSDR RF output is plugged to U-blox GPS receiver through 40dB attenuator. (U-blox receiver has antenna power bias option disabled). USB port of GPS receiver is connected to another PC running U-Center on Windows 10.

I also tested with some mobile phones and they can't fix the location.

[Mictronics]

Feeding from laptop should be a problem for the IQ stream. However, with a direct feed your gain setting 0dBm seems to be way to high. Keep in mind, the guaranteed signal strength on earth for GPS is -130dBm. Even with a 40dB attenuator, your GPS still "see's" -40dBm. So you overpower the GPS low noise amplifier, cuttings its sensitivity and dynamic range.. See my screenshot, I use -40dB with pluto's oem antennas and the GPS in 2m distance. SNR is still 26dB, can go down to -50dB and still have a SNR of about 8-10dB which is enough for proper reception. Keep in mind, real GPS is a very week signal, that's what the receiver is made for.

[ferorted]

Sorry, the images with att=0dB was set only for spectrum analyzer measurements. When I did the tests with PlutoSDR feeding the GPS receiver, ATT is from -60 to -80 (+40 external attenuator), total: -100 to -120 dBm aprox.

Anyway, if the problem was that the IQ stream from host to Pluto, it would also affect the testing with the "master" code, however with "master" code the spectrum analyzer shows a good signal.

[Mictronics]

Have you tried the GPS outside to check whether it still works? Tried a different GPS module? Just with air cap, no direct connection?

[ferorted]

The GPS receiver works fine with antenna and real GPS signals.

I also did some tests with antennas (without LNA) in Pluto and GPS receiver.

I'm going to try to get another Pluto for testing.

But the readings on spectrum analyzer with the last testing code are very strange.

El lun., 17 feb. 2020 20:52, Mictronics notifications@github.com escribió:

Have you tried the GPS outside to check whether it still works? Tried a different GPS module? Just with air cap, no direct connection?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/Mictronics/pluto-gps-sim/issues/6?email_source=notifications&email_token=AOBCA5Y6QEYY52W6RXDUCU3RDLTI3A5CNFSM4KT67DI2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEL7QABI#issuecomment-587137029, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOBCA55UTNXGV2WHPGGGHNTRDLTI3ANCNFSM4KT67DIQ .

[ferorted]

Hi again, For your reference, I did some throughput tests by the USB port. As you can see in the images, the speed trafic with the "master" code (first image) is about 11.000 packets/sec and constant. With "testing" code (second image) , the speed is only about 8.800 packets/sec and is less constant than previous.

These testing were done with laptop to Pluto USB port, while pluto was powered by its Power-USB port with an external 5v/3A adaptor.

P.D.: tomorrow I hope getting another Pluto device and testing with the GPS receiver again. Best regards.

[ferorted]

Hi, finally I could tried a new Pluto SDR and it WORKS fine !!! It's incredible but my old Pluto doesn't work property with pluto-gps-sim code. Anyway, your "testing" code neither works with my new PlutoSDR.

Now i'm going to testing with some mobile phones and pluto-gps-sim.

Thank you very much.

[Mictronics]

Did you do anything special before with the old Pluto? I there a chance that the TX output gain stage is blown? Strange anyway.

"testing" code neither works with my new PlutoSDR

What does this mean? Same behavior, different, low power?

[ferorted]

Did you do anything special before with the old Pluto? I there a chance that the TX output gain stage is blown? Strange anyway. --> Nothing, Tx provide good output power (as you can see in previous spectrum analyzer images). Maybe the problem could be the precision of oscillator, but i can't test it.

"testing" code neither works with my new PlutoSDR What does this mean? Same behavior, different, low power? --> With the "testing code" the problem is the same than previous Pluto, the RF signal is not constant. Please, see my previous short video "branch-testing"

[ferorted]

The only difference is old Pluto is version 0.31 (I updated it before testing) and new Pluto is 0.30 But now i'm afraid to update this :-)

[Mictronics]

I run 0.31 as well on mine.

[ferorted]

Hello, I'm not a lucky man :-( Now my GPS receiver get generate signal by PlutoSDR and fix the location perfectly. But none mobile phone can fix the location. In these images the only estrange thing is the showed Pseudo Range in uCenter, because there are differences between received and the generated. Is it correct? Any suggestion about which parameters to check for ensure it proper works with phones? Any other idea?

[Mictronics]

In case your phone is on Android: Get the app sat-stats and check what other satellites are visible to the phone.

My experience with a Samsung Galaxy Tab 2: sat-stats is showing me exactly the PRN's I simulate for GPS. However, the tablet receives in parallel Glonass satellites or Beidu. As a result, the data from simulation and from other sources does not correlate, hence the tablet never gets a position fix. So as long as you can not exclude other sources for position data, and isolate the GPS, you will not succeed with a mobile device.

[ferorted]

Thanks, I'm going to test with app sat-stats. But my tests are done in a place without real GPS signal reception.

[ferorted]

Hi, I did many tests with the sat app and my phone (Huawei P9) but could never fix the location. Additionally in the last test I used the Huawei P9 phone and also another old ZTE (stored some months ago), it has Internet wifi access but not SIM card. Amazing, the ZTE fix the location !!!!!!!!

I did more tests changing Pluto output power, distances ... everything was ok.

[ferorted]

[ferorted]

So, I switched off the Pluto, put a SIM card into the ZTE, went outside to clear sky and the phone took a long time to fix the real location.

Then I did the same simulated tests than previous (without SIM neither real GPS signal), and... the phone didn't fix the location !!!

I think when the phone get a real GPS signal and it is synchronized it is hard to spoof with other signals.

[Mictronics]

Excellent work. But your result proofs what is expected: The location algorithm on modern mobile devices are hard to spoof.