Enhancements to Dump978 (UAT decoder)

Well, I did update and there where zero messages recorded, waited a few hours.
Now I reverted to mutability branch and immediately messages appeared. No planes are visible as of yet.
Should I mention that I am using dump1090-fa?

LE: Haha, this was unexpected…

I am using dump1090-fa (Piaware SD card image), and have installed dump978 yesterday using JP’s script. Zero messages here too. Will try dump978 on Raspbian Stretch + dump1090-mutability.

Fresh install.

STEP-1:

Burned microSD card with Piaware SD card img 3.5.3
Booted and checked folder /var/www/html/

pi@piaware:~$ ls /var/www/html
index.html index.js index.lighttpd.html lib logo.gif translations

STEP-2:

Ran JP’s script and installed ONLY dump978.
Checked, MAP NOT showing.
Checked folders. The folder dump978 is created and contains folder data, BUT NO MAP FILES

pi@piaware:~$ ls /var/www/html
dump978  index.js   lib  translations
index.html  index.lighttpd.html  logo.gif

pi@piaware:~$ ls /var/www/html/dump978
data

pi@piaware:~$ ls /var/www/html/dump978/data
aircraft.json  receiver.json

STEP-3:

Again ran JP’s script and installed Web Portal.
Checked again, all the necessary files for MAP are now available

pi@piaware:~/adsb-receiver$ ls /var/www/html
admin     dump1090.php  graphs.php  lib         templates
api       dump978       index.html  logo.gif    translations
blog.php  dump978.php   index.js    plot.php
classes   flights.php   index.php   post.php
data      graphs        install     system.php


pi@piaware:~/adsb-receiver$ ls /var/www/html/dump978
config.js  dbloader.js   gmap.html   ol3             style.css
coolclock  flags.js      jquery      planeObject.js
data       flags-tiny    layers.js   script.js
db         formatter.js  markers.js  spinny.gif


pi@piaware:~/adsb-receiver$ ls /var/www/html/dump978/data
aircraft.json  receiver.json

STEP-4:

(1) Configured Web Portal by going to page IP-OF-PI/install

(2) Edited file config.js and changed SiteShow = false; to SiteShow = true;

 pi@piaware:~$ sudo nano /var/www/html/dump978/config.js

(3) Edited file dump978-maint.sh and changed device number from -d 0 to -d 1

pi@piaware:~$ sudo nano adsb-receiver/build/dump978/dump978-maint.sh

#!/bin/bash

# Start dump978 without logging.
while true; do
    rtl_sdr -d 0 -f 978000000 -s 2083334 -g 48 - | /home/pi/adsb-receiver/build/dump978/dump978/dump978 | tee >(/home/pi/adsb-receiver/build/dump978/dump978/uat2json /var/www/html/dump978/data) | /home/pi/adsb-receiver/build/dump978/dump978/uat2esnt | /bin/nc -q1 127.0.0.1 30001
    sleep 15
done

STEP-5:

(1) Rebooted Pi
(2) Checked Map, available at IP-OF-PI/dump978.php
(3) Checked Range Rings, available now.

1 Like

@abcd567

Are you using my branch? If so can you see RSSI measurements in the dump978 map? It’s working great here as seen below. These even come through to the dump1090-fa map.

1 Like

No, I am still at JP’s.
The problem is that there is zero UAT traffic here (Toronto) since yesterday (when I installed dump978), so I cannot judge how it works, and if I go to your branch, how it will improve.

1 Like

I got the same: Circles correctly located at my location, but Station Marker south tip of South America :slightly_smiling_face:

dump978-station%20marker%20off

I solved it by following method:

pi@piaware:~$ sudo nano /var/www/html/dump978/config.js

Scrolled down, and removed quote marks " " around numeric values of SiteLat and SiteLon

BEFORE EDITING

SiteShow    = true;           // true to show a center marker
SiteLat     = "43.xxxx"
SiteLon     = "-79.xxxx"
SiteName    = "My Radar Site"; // tooltip of the marker

AFTER EDITING

SiteShow    = true;           // true to show a center marker
SiteLat     = 43.xxxx
SiteLon     = -79.xxxx
SiteName    = "My Radar Site"; // tooltip of the marker

Now marker flew back from South Pole to Toronto. :smile:

dump978-station%20marker%20on

1 Like

You are trying to run on device 0! Are you sure it is that one? In my case 0 is for 1090, 1 is for 978…

How do you figure things out?? That’s great, it worked thanks!

Now, it might be true that being Sunday, there are not so many UAT flights here either, so I won’t give up yet.

I have edited the /adsb-receiver/bash/decoders/dump978.sh - edited the lines:
147
git clone GitHub - KeyserSoze1/dump978: Experimental demodulator/decoder for UAT on 978MHz

292
tee >(${RECEIVER_BUILD_DIRECTORY}/dump978/dump978/uat2json --rec-pos lat,lon ${LIGHTTPD_DOCUMENT_ROOT_DIRECTORY}/dump978/data)

I have used the internal MC editor because nano doesn’t have line numbering.

Then back to re-install the dupm978 with JP’s script:

    cd ~/adsb-receiver
    git pull
    sudo ./install.sh

After that, back to changing the device for dump978 to device #1
sudo nano /home/pi/adsb-receiver/build/dump978/dump978-maint.sh

The script posted showed the default settings. I changed -d 0 to -d 1 in my Pi. I have mentioned this in that post.

I learned this trick from Professor Dumbledore at Hogwarts School. :slight_smile:

Shoot, I just heard a small plane above. Nothing showing in 1080 ADS-B, nothing in 978 UAT… not even messages.
LE: talked too soon - victory!

Now I need a better antenna for 978MHz. Right now I have a Pepsi can with a paper clip sticking up.
Hmmm… I wonder what’s the bandwidth of the FA 1090 antenna?

1 Like

From this page:

https://www.rtl-sdr.com/review-flightaware-ads-b-antenna-and-filter/

I quote the following:

“In the SWR test both antennas had excellent SWR at around 1090 MHz, with both Adam’s antenna and the FlightAware antenna being about 1.15. The FlightAware antenna however has pretty poor SWR of ~10 at 978 MHz, so it may not be so good for UAT reception at 978 MHz. Adam’s antenna has a much better SWR of 2.0 at 978 MHz.”

It does not answer your question directly, but it’s likely enough info, if your plan was to use the FA antenna for UAT.

Exactly that’s my plan. Will think about it… how to splice the antenna output without loses.

Whe I was experimenting with receivers and filters and amplifiers and I wanted to ensure I was comparing the same signal into each system I used one of these

Two-way Low Power Splitter, 555–2700 MHz

The one I bought new for $AU5 at a hamfest is branded ANDREW with the same part number.

Insertion Loss at Frequency Band 0.3 dB @ 555–960 MHz | 0.4 dB @ 1710–2700 MHz

I tried using Satellite splitters and combiners with some success but the ANDREWs one is really good.

S.

Without amplification those loss figures are impossible. Most likely that is in addition to the 3dB from splitting the signal in half. So that makes the loss figures 3.3dB and 3.4dB. Minicircuits quotes their loss figures the same way. Still good though for a 2-way splitter.

I have a preamp that I can use. One transistor, gain not too high if supply power is 3.3V.
I might not even need it…

I just updated.
downloaded the files to a new dir, did a make and copied the executable files over the existing ones.
I now get RSSI. Will make gain optimisation so much easier.

I have one RTL-SDR connected to a DPD antenna with the hab/nevis amp powered by bias-t(Hardware mod). The other one is the FA pro stick(without the filter) feeding a directional skymaster antenna.
Range is limited as I am low and UAT978 aircraft tend to fly low.

1 Like

Great to hear that it seems to be working. Gain is tricky as they can go from weak to strong very quickly due to the elevations most of the aircraft fly at. I can get some that are 150 miles away but those are ones that are up at 30k+.

I have a small wifi like antenna designed for 978 (similar to Nooelec - 978MHz UAT ADS-B Antenna, 5dBi (High Gain), SMA) sitting about 20 feet off the ground running to a FA filter and then a FA prostick. I’ve found upper 30’s in gain seem to be best for that setup but it’s pretty variable depending on the level of activity.

With the amount of traffic I know there aren’t a lot of people using this but I got tired of not being able to easily tell if the signals were weak or strong and decided to add it. I hope others can get some use from it.

1 Like

I am still waiting for any aircraft with UAT978, but alas none since I have installed dump978 two days ago

pi@raspberrypi:/var/www/html/dump978/data $ cat receiver.json  
{
  "version" : "dump978-uat2json",
  "refresh" : 1000,
  "history" : 0
}
pi@raspberrypi:/var/www/html/dump978/data $ cat aircraft.json
{
  "now" : 1525111716,
  "messages" : 0,
  "aircraft" : [

  ]
}

LATER EDIT:
could not understatnd "now" : 1525111716,, so did a conversion

pi@raspberrypi:/var/www/html/dump978/data $ date -d @1525111716
Mon Apr 30 14:08:36 EDT 2018
1 Like

That’s correct, living next to a big airport you won’t see too much UAT traffic, they usually can’t afford the fees. The small planes tend to go to smaller airports, and because they fly so low, it’s harder to “see” them.
Plus, I see a lot of small planes being equipped with ADS-B, bypassing UAT completely.

1 Like

I have seen posts in the past about NOT using AGC.
One user tested it and found it worked well for them(on 1090Mhz) as they are in a quiet location.
I live under the KLGA arrival path and next to the Hudson River SFRA so get way too much traffic to do this.

978UAT is another thing. It is often quiet, especially when the weather is bad.
I thought I would give AGC a go. There are no secondary radar responses so I thought it may help.
It seems to work well. My range seems to have improved a little. It has only been one day so I will have to let it go for a while longer before making any conclusions.

It’ll depend on the installation. In my case, if I use AGC, I get a single line between 0 and -3 dbfs, and no decodes whatsoever.

On the other hand, if I lower the gain, there are some very weak decodes. I tried 20.7 again yesterday. The noise line was around -35 dbfs. I saw a plane with a -49.5 signal. I’m inclined to believe it was not real, but it tracked okay for a period of time. Past posts here on the forum mentioned that the dongles/sticks are good up to -30 dbfs.

Anybody has more info about how ‘low’ will they go?