FlightAware Discussions

HOWTO: Airspy mini and Airspy R2: Piaware / dump1090-fa configuration

You cut some of the numbers…

How do I make just a graph bigger (not all of them)? Like this one…

uh, i cut the left part as the graphs were empty. That’s the full chart

1 Like

There are numbers there on the bottom.

not sure what you are still missing. That’s the full image

I meant that the first image was missing those numbers. Second one has them.

Ok, seem to be we both replied the same time.

1 Like

The biggest difference is that you can run with higher -e than a Pi4 and with less risk of dropping samples when the CPU usage is higher. With the later versions of airspy_adsb I doubt that there are any benefits with higher -e since the curve seem to drop of after a certain level. I’m going to do some testing of that when I get a new Pi4. I will connect it to the same antenna as my PC and then compare stats.

If I hadn’t had this PC laying around I wouldn’t have invested in one for running an ADS-B station. I’ve put in some time and money on the signal chain (good antenna, cavity filter, reasonably decent coax and LNA) and felt that any improvement that could be had was by adding more CPU and better USB subsystem. It was perhaps correct, but the added electricity the PC requires makes the cost hard to justify (for me). So I’ll probably end up running on a Pi4 again once I’m done with my testing.

2 Likes

Same here, I had one of my older laptops (with an i7-2640M CPU @ 2.80GHz) laying around, not being used anymore. That has 4 virtual cores, not that it would matter, airspy_adsb uses just one.

My chart is different again: probably a fluke but as you can see, it’s difficult to distinguish DF4 from DF11…

df_counts-localhost-2h

… when the totals are very similar.

Is there a way of configuring the colours / overriding the default choice?
Just a thought.

1 Like

My pi 4 is overclocked to 2 GHz, so I have some reserve computing capacity in case. Highest available preamble filter level is already 21 with -C 85 even at night.
I’ll step towards the solution of using a PC if I find a good offer for a used one. Maybe not for moving ads-b decoding there, but for my other projects. Now, my experimental projects are on my main pc, in VirtualBox/Ubuntu (Linux Mint).
RF chain is ok here, except the antenna position. I have to get permission to move the antenna outside to the roof. Now, it is in the attic, under tiles - with moderate receiving circumstances.

If I decide to buy a used pc, I would rather pick one with high core frequency instead the one with more cores - for the same reason.

So maybe A2A DF0 are mostly generated by smaller/GA aircraft, or when other aircraft are in close proximity? Not seeing DF0s overnight, until about the time small aircraft start to fill the sky in the morning.
df_counts-localhost-8h

I think DF0 goes up when the traffic gets busy and planes get closer. So the planes get more “cautions” and start interrogating their neighbors.

Understanding Mode S technology – a discussion about Mode S basic, elementary and enhanced surveillance, DF17 extended squitter and ADS-B | Emerald Insight

UF is a specific interrogation originating from SSR or another airframe asking specific, addressable information about that airframe. DF is the reply from the airframe in regards to the UF interrogation. UF 0, 4, 5, 11 and 16 make up basic surveillance. Basic surveillance messages are made up of the airframe address, parity bits and a 56-bit data word known as “short” interrogations and replies.

UF0 is a short air to air surveillance for TCAS/ ACAS. The DF0 reply will have in it the Mode C altitude as well as the Mode S address. To test DF0, its altitude reply is compared to the Mode C altitude and the Mode S address for verification. Also encoded in the DF0 reply is the vertical status (VS) bit, the reply information (RI) field. The VS bit will indicate 1 if the airframe is on the ground and 0 if it is on the air. The RI field is a 4-bit word containing the airframe’s true speed capability and type of reply to the interrogating airframes.

1 Like

OCD Friendly DF
Screenshot_249

1 Like

@prog @wiedehopf
Just had an error in log. It seems to be handled, but a dev is needed to find out if there is something to do. (month might look strange for you - it is about localization) Max cpu temp was just below 50°C
It is interesting that preamble filter started with the latest used value (25) instead the default (4).

  • I pick a lower value for -C like 80 to see if it is about cpu load. I limited also the preamble filter -E 16 (Now adsb cpu is 10% below the target.)
szept 15 15:52:29 raspberrypi airspy_adsb[30837]: airspy_stop_rx() failed: AIRSPY_ERROR_LIBUSB (-1000)
szept 15 15:52:29 raspberrypi airspy_adsb[30837]: Decoding stopped
szept 15 15:52:29 raspberrypi airspy_adsb[30837]: Push client disconnected from localhost:30004 (beast)
szept 15 15:52:30 raspberrypi airspy_adsb[30837]: Acquired Airspy device with serial 056C68DC2145A09F
szept 15 15:52:30 raspberrypi airspy_adsb[30837]: Decoding started at 12 MSPS (Gain: auto; Preamble Filter: 25)
szept 15 15:52:42 raspberrypi airspy_adsb[30837]: Push client connected to localhost:30004 (beast)
szept 15 15:53:10 raspberrypi airspy_adsb[30837]: airspy_stop_rx() failed: AIRSPY_ERROR_LIBUSB (-1000)
szept 15 15:53:10 raspberrypi airspy_adsb[30837]: Decoding stopped
szept 15 15:53:10 raspberrypi airspy_adsb[30837]: Push client disconnected from localhost:30004 (beast)
szept 15 15:53:11 raspberrypi airspy_adsb[30837]: Acquired Airspy device with serial 056C68DC2145A09F
szept 15 15:53:11 raspberrypi airspy_adsb[30837]: Decoding started at 12 MSPS (Gain: auto; Preamble Filter: 25)
szept 15 15:53:23 raspberrypi airspy_adsb[30837]: Push client connected to localhost:30004 (beast)
szept 15 17:42:45 raspberrypi airspy_adsb[30837]: CPU 90.0 %, target 85.0 %, adjusting preamble filter: 25 -> 24
szept 15 19:12:37 raspberrypi airspy_adsb[30837]: CPU 90.5 %, target 85.0 %, adjusting preamble filter: 24 -> 23

Just checked the MTG box and saw this in the log.

Sep 15 17:13:42 g0pkt-piaware airspy_adsb[31142]: No data from device for 15 seconds, exiting!
Sep 15 17:13:42 g0pkt-piaware airspy_adsb[31142]: Apparently the data processing didn't finish, report this please.
Sep 15 17:13:42 g0pkt-piaware airspy_adsb[31142]: Decoding stopped
Sep 15 17:13:42 g0pkt-piaware airspy_adsb[31142]: Client disconnected 127.0.0.1:55656 (beast)
Sep 15 17:13:58 g0pkt-piaware systemd[1]: airspy_adsb.service: Service hold-off time over, scheduling restart.
Sep 15 17:13:58 g0pkt-piaware systemd[1]: Stopped Airspy ADS-B receiver.
Sep 15 17:13:58 g0pkt-piaware systemd[1]: Started Airspy ADS-B receiver.
Sep 15 17:13:58 g0pkt-piaware airspy_adsb[31139]: airspy_adsb v2.2-RC13
Sep 15 17:13:58 g0pkt-piaware airspy_adsb[31139]: Listening for beast clients on port 47787
Sep 15 17:13:58 g0pkt-piaware airspy_adsb[31139]: Acquired Airspy device with serial 26A464DC287E3993
Sep 15 17:13:58 g0pkt-piaware airspy_adsb[31139]: Decoding started at 12 MSPS (Gain: auto; Preamble Filter: 1)

So I’m reporting it @prog @wiedehopf

The latest is RC14 with brand-new graphs1090 curves
(DF distribution of decoded messages)

You have a couple more lines above that for good measure?

Unstable USB connection. (or more precisely the airspy stopped streaming for some reason and then the orderly shutdown command didn’t work).
Was that on boot by chance?