Graphs for dump1090 -- my version with install script

GitHub - wiedehopf/graphs1090: Graphs for readsb / dump1090-fa / dump1090 (based on dump1090-tools by mutability)

It says install / update, there would be a note if it deleted data, believe me.

Nothing of consequence has been changed in months.

cd /usr/share/graphs1090/git
git log

This will show you the installed version, you can compare with the git commits on github.

1 Like

I’ve readded the noise reading to the signal graph, might be useful especially when you have basically no traffic and the weakest signal isn’t representative.

Only works with a dump1090-fa/readsb working directly with an rtl-sdr receiver.

1 Like

Was drawing the same noise line on the UAT signal graph an intentional add? I suppose it is interesting for 978 MHz as well even if it comes from a 1090 MHz dongle reading.

(I don’t currently have UAT aircraft tracking on my test station due to smokemageddon, but I noticed the noise line looks identical to the ADS-B/dump1090-fa graph. I’m not implying that I mind.)

chrome_agvLZEGvEw

chrome_G8LCCXDJ33

No it was not, i’ll fix it.

Edit: should be fixed.

Good morning… the line is gone now (UAT). :slight_smile:

Hello @wiedehopf! I’m new here on the discussion list, and newbie using graphs1090 (Congratulation - great job!). I’ve a basic question: please, could you indicate some link, post or article that explain the graphs of graphs1090? The axis units, differences between tracks, messages, aircrafts and other. Thank you!

I don’t know if there is such an article to link to. Anyway, here’s a basic explanation of what each graph shows:

Message Rates:

This graph shows an instant count of how many messages your receiver is decoding from all aircraft. The dark blue line is how many messages of all types per second are being received. The green line is the average of this count for the preceding 7 days. The yellow band indicates the maximum and minimum values over the preceding 7 days. The current rate will usually be somewhere near to this band, and if it’s far outside it can indicate a change in receiver performance assuming traffic has been fairly constant. eg damage to the antenna.

The light blue line indicates the number of positions of aircraft being received per second. The red line will appear only if you are using an rtl-sdr dongle and indicates what percentage of the signals received have a very high signal strength. This can be useful for tuning the gain of the receiver - if it’s too high it can become overloaded by close by aircraft leading to messages being missed. If this value is higher than about 5% consider reducing gain slightly.

Aircraft Seen:

This graph indicates the number of actual aircraft your receiver is seeing. The solid green area indicates the total number of aircraft. The blue line is the number of those that are transmitting their position via ADS-B. The black line is the number of aircraft who are not transmitting their position, but for whom a position has been calculated using multilateration.

The red line shows the number of aircraft received that are not transmitting their position, and no position has been calculated for them (perhaps because there are insufficient other receivers in range to do so). This number will also include a few bogus decodes resulting from noise and aren’t actually there. It will also include aircraft that get an mlat position later on.

The yellow line shows the number of messages received via TIS-B, which is those relayed by a ground station and can include aircraft being tracked by ground radar and not ADS-B. It’s only relevant in the US as that system isn’t used elsewhere.

Aircraft tracks:

This graph is probably the least useful - it shows the rate at which new aircraft are being detected by your receiver. The red line shows aircraft tracks where only one message was received - this is highly likely to not have been a real aircraft, but a bad decode from a corrupted message. The less red the better.

Range:

This graph shows a summary of the ranges of the aircraft you are receiving. The dark blue line is probably the most interesting as it indicates the maximum range you are receiving. The higher this is the better, but it will be limited by the terrain surrounding your receiver.

The black line is the median range received, and the green band indicates the upper and lower quartiles. The light blue line is the minimum range received. These are less useful for assessing receiver performance since they are mostly affect by where the aircraft you receive are flying.

Maxima:

This graph shows the absolute maximum values for each line shown. When you change to a graph for a longer time period (a year for example), the resolution of the graph is reduced because of how the data is stored. This means that the other graphs containing those fields will not reflect the peak values as well, but more of an average.

The fields should be fairly self explanatory, but the message rate graph is split into ‘local’ and ‘remote’. Local messages are those received and decoded using dump1090 directly, eg if you have an rtl-sdr dongle connected. Remote messages are those which are fed to dump1090 via a network connection. In my example, you can see I only have ‘remote’ messages as I’m using an airspy, not an rtl dongle.

Messages per aircraft:

This is fairly self-explanatory. The blue line is simply the number of messages received per second, divided by the number of aircraft being received. ADS-B is a ‘non-cooperative’ protocol, in that aircraft do not listen to see if anyone else is transmitting before sending their own messages. This means that when many aircraft are nearby, they transmit over the top of each other and some messages will be lost. The higher this number the better, but it’s of limited use for assessing receiver performance since its affected so much by the amount of traffic and settings used by interrogating radar stations.

The remaining graphs are just standard system monitoring and aren’t specific to ads-b.

15 Likes

What does noise level really mean? I’m a little confused why weakest ADSB signal can be below noise level. I thought signal can’t be picked up below noise level.

image

All received power that is not part of a successfully decoded message.

It is not the noise floor.

(This “noise” calculation is a relatively cheap calculation that can be done continuously on all received samples. It might be interesting to look at periodically taking some of the incoming samples - 1% of samples or something like that - and doing a proper analysis of that, on the assumption that the data looks mostly the same over time)

3 Likes

This was great, thank you!

Excellent :+1: :clap:

1 Like

Caius the Pioneer

@caius was probably the first one among us to use collectd and generate graphs using @obj scripts at Github. He first reported/published this in April 2015.

Eight months later in December 2015 J Prochazka published his famous " ADS-B Receiver Project Setup Scripts" which installed, among other things, graphs using @obj’s scripts at Github.

 

Below are links to @caius’s three posts, dated April 12, 2015, giving details about graphs

  1. https://forum.planefinder.net/threads/ads-b-diy-antenna.23/page-135#post-3978

  2. https://forum.planefinder.net/threads/ads-b-diy-antenna.23/page-136#post-3992

  3. https://forum.planefinder.net/threads/ads-b-diy-antenna.23/page-136#post-4010

 

4 Likes

Doesn’t time fly…

2 Likes

@caius thank you! Great post. I’ll use it to understand the graphs generated from my new station here in São Paulo, Brazil. It’s a recent installation, then I’ll need to adjust the gain, test the effect of different length of coax, the height of the antenna, and other parameters.

Thanks!

@caius and others: I’ve two stupid questions:

  1. What’s the difference between “tracks”, “messages” and “positions”? Can I think in:
    [Aircraft 1]
    [Track 1]
    [Message/Positon]1
    [Message/Positon]2

    [Message/Positon]n
    If yes, messages = positions?

  2. The term “RHS” on the graphs legends means some like “[R]oot [H] [ -something- ] [S] Square”, like RMS?

Thanks!

“Right Hand Side”. Alternative y axis scale.

2 Likes

Messages are individual transmissions from aircraft. Some message types include aircraft position, but not all. There are several different types of message that carry various items of data such as altitude, airspeed, navigational status and autopilot settings among other things. Some of these messages are transmitted automatically on a regular basis (something like twice per second for ADS-B messages), and some are transmitted in response to an interrogation from a ground based radar. If you click on an aircraft in dump1090, you will get an idea of the type of information transmitted in the detail pane.

The more aircraft in range, the more messages you will receive and if you live in an area with many ground radars that actively query aircraft you will see a lot more messages.

In Europe, radars tend to request more data from aircraft than they do in the US for example, so a receiver here will probably see many more messages per second for the same number of visible aircraft.

The tracks graph is actually a rate - tracks per hour. It is showing how many new tracks the receiver is seeing per hour. It’s not the same as aircraft, which is a count of the number of aircraft visible at that time. An aircraft is counted as a new track when it’s first seen, so if a plane flies into range and stays visible for several hours, it would count as one track. If however, it came into range, then dropped out of range again (or went behind some terrain for a while) then came back into range again, and did this several times, it would count as a new track each time. There is a minimum time it has to be out of sight before it’s counted again. I’m not sure what it is, but it’s of the order of a few minutes.

1 Like

Thanks @caius and @obj for the information.

As I said, recently (1 week) I installed my station on the rooftop. My configuration is: RTL-SDR RTL2832u dongle + Raspberry Pi Model B+ v1.2 (2014) running Linux raspberrypi 5.4.51+ #1333, dump1090-mutability 1.15, 75 ohms coax cable with 1.5 m, and 15 m ethernet CAT5 cable from station to my router. I don’t have LNA amp, and the antenna is a dipole of 1/4 wave (“8 legs” spider). Some pictures here. I’ve been a lot of adjusts, testing different ppm, gain, and other configurations to obtain max efficiency. Fell free to make some suggestions!

Thanks!


2 Likes

This is unmaintained and you probably want to upgrade to something supported e.g. dump1090-fa.

You seem to have open wires between the F connector and your dongle.
Also looking at your antenna you again seem to have open wires.
You need to use co-ax cable.
You will be picking up lots of noise on these open wires, as well as having severely mismatched impedances.

2 Likes