Some Speed Tests with Feeder

Hey everyone,

because of the use of an umts stick in combination with a pi, I was interested how many kb/s are necessarily to send data with the opensky-feeder orderly.

So I made a little 30 minute test shown on the picture.



Cheers,

Andy

Always wanted to do this

I just had a look at the bandwidth usage of each sensor on our servers. Our strongest receiver is producing 2000-2500 messages per second (peak) and generates inbound traffic of about 60-70 kByte/s. So it's pretty much the same, although it's a Radarcape. But also that's not surprising since both use the exact same network protocol (Beast protocol) to send data to our servers

This data rate is interesting for cost-efficient use of a UMTS stick. Many phone company contracts allow a data rate of 64 kBit/s after the high-speed data volume of the booked package is exhausted.

Hi....I will look into this but I cannot see why an antenna would need to have a DC short. There are 2 connection points - signal and ground. If you short the signal to ground, the RF signal will be shorted straight to ground and will not function correctly, in the case of a passive antenna. The exception may come where one uses an antenna which has gain circuitry.

seo consultant

Hi there....

the question is ....about what we are talking because of reception.
@Schaefer
"Our strongest receiver is producing 2000-2500 messages per second (peak) "


what does this mean exactly ? Are this complete messages overlapping messages....or evaluated messages wit crc check ?

Are this messages in total in the air (receivable) or is the receiver limited factor.

If the receiver is not the limited factor the messages per second should be in the range of 10k ! or more.

For radarcape it should not be an issue

I do not think that 30kf/s serial in a row (no overlapping) are a problem for radarcape .... should not be an problem

regards

BILL2002

Hi BILL2002,

Are this messages in total in the air (receivable) or is the receiver limited factor.

That's the number of correctly received messages with valid crc check.

As you stated above the theoretical upper limit it way higher than 2,5k messages/s. However, ADSB or MODE-S, respectively, use random medium access. Transponders can't know for sure if the channel is free for all possible locations of reception (hidden node problem). At first one might think an antenna that is located at a very high ground with perfect line of sight in all directions for hundreds of kilometres should get the best results w.r.t. the number of received messages. It is certainly true that the antenna will receive signals from more aircraft than an antenna that is located on a lower ground. However, the chance of receiving overlapping messages increases with the number of aircraft and at some point the message rate will drop again due to these collisions. A Radarcape can't change the fact that overlapping messages can't be evaluated correctly and therefore is capped at ~2500k, too.

Best,
Marco

I understand....
Beast is able to receive > 2.5k f/s
So question what is the frame rate of a SDR dongle together with a PI for feeding application ?
As I can see in the picture of achan011 the framerate is below 200 f/s
Do you have any idea ?

cheers

Bill2002

I don't know exactly what his setup is, but it seems walleyv (currently holding rank 1 of the average messages toplist ) is using a dump1090 decoder (on a rPi?). He's currently getting ~3500 messages/s:

opensky-network.org/receiver-profile?s=333333990

That's pretty insane. I'm guessing he's combining multiple SDR dongles, but I don't know.

thebaldgeek (currently holding rank 5 of the average messages toplist ) had a messagerate of ~2500 earlier today , also running a dump1090, but again I don't have any details on what hardware he is using.

Cheers,
Marco

I don´t think that this is a SDR setup together with a PI
You are right he is merging 2 receivers.

So come back to the basics ....
what is the real framerate you can get with a SDR and a PI ? in average ?
Any suggestions ?

rgrds

Bill2002

what is the real framerate you can get with a SDR and a PI ? in average ?

I don't know and I'm not sure if anybody knows for sure, because that's not an easy question to answer. The framerate depends on a lot of different factors and not being able to improve it any further doesn't mean it can't be done by somebody else with a different antenna and/or location. Maybe some radio enthusiast, who has played around a lot, will stumble over this topic some time and provide you with some hands on information.

Cheers,
Marco

Thank you Marco,

I imagine that you folks have made a test in the past

So I will do that test by myself. I think that the Raspi together with a SDR should be very limited by the processing power of the pi.
And SDR will have prob. with the AGC because to slow.

Have a nice day

Bill2002