How Do I Optimize Gain?

This has been a fun few days for me. First I received the new FlightAware dongle, which immediately improved my reception by around 50%. Then, I moved my center loaded whip antenna from the window sill inside to the window sill outside. My numbers have more than doubled!

All of a sudden I’m seeing planes to-and-from Hawaii and the South Pacific for a few hundred miles.

I am curious if my receiver’s gain is optimized for this new set-up. Here are my performance parameters: … hoto%2cpng Click “View original” to get it large enough to read.

Any advice will be appreciated.

Hi Geoff,
Your results look pretty sweet to me for a window/indoor set up with range of a few hundred miles.
Great location by the look of things on your stats.
Moving your antenna outdoors would get even better results.

I haven’t been able to get a clear answer to this question either. But following this formula has proven fairly successful for refining my gain thus far. The goal is to keep messages stronger than -3 dBFS below 3 percent…

uhh, I guess I should dial mine back a bit after installing the ProStick :open_mouth:

I’m not sure if that formula for “messages over 3dBFS” holds true for every site.

I’m right in the heart of some very busy NYC airspace (about a mile and a quarter from runway 4 at LGA) and I get way better range, message rate and ac track when that ratio is about 6 for me than when it’s 2 or 3. But - I’ve got a hell of a lot of aircraft moving low and slow within spitting distance of my antenna.

I think it’s a good rule of thumb to start from - and obviously obj knows what he’s talking about - but ultimately remember that you’re seeking to maximize message rate, range and a/c tracked. If you dial back your gain and your other numbers plummet - don’t take the listed ratio as the golden rule.

Best answer I can give is change the gain one step at a time and monitor both message rate, and number of planes on the dump1090 screen. If you are running dump1090-mutability and have the collectd graphs enabled, change the gain one step at a time and monitor the changes on the graphs. Generally speaking, if the change results in higher message rate and more planes at the same time, it is a better gain setting.


Thanks everyone for your help. Since the signals are constantly changing it will take a while before I know if what I did will produce better results,. My gain is now set at 37. I am using the FlightAware filter between the FA dongle and whip antenna.

BTW – my antenna is on the outside window sill. It has a pretty open path to the west. Though it’s too far (12 miles) to say I have an ocean view, at sunset I often see sun glint off the ocean between the coast and Santa Catalina Island (which I also see). I am consistently tracking planes to ~240 miles to the west which is about the theoretical horizon for planes at altitude.

So far, so good.

Now that the gain has been tweaked I thought I’d report back.

It went in two steps from -10 to its current 37. I’m not sure how gain is calibrated or if its calibrated at all. Its just the spot I found.

I have a nearly unobstructed view of the Pacific 12 miles away. Seeing Hawaiian, Australian and New Zealand bound planes came when I moved the antenna outside on the window sill. There is no loss of distance with this lower gain. My location receives these planes almost as soon as they pop up over the horizon… sometimes actually farther.

  1. -3dBFS hits way down. I’m not sure why that’s the magic number, but it seems to be.

  2. First time ever, my noise floor is a variable. It had always been mostly rock steady. Something was probably being clipped before?
  3. Using less cpu
  4. Everything else seems about the same.

My goal is to optimize. At this point additional gains will be small. This might pull a few additional weak signals, but my limitations are more topographical.

Results today are at least four or five times the traffic when I first set things up. It’s all been in small increments until I moved the antenna to the outside.

A huge thank you to all the volunteers who’ve worked so hard especially Joe, who distro I use.

Hey everyone - I’m running a home made spider antenna on a N-Type bulkhead connector. I picked up a ProStick and FA filter.
At first, I had little 12" jumper, to go Type-N to SMA.
Then I bought a L-Com Type-N to SMA adapter. Thinking if I removed any addt’l length of wire - the better.
I’ve messed with my gain (when using the jumper), and I’m thinking from what I’ve read, my value, 37, is working out best - but I figured I’d let you tell me.

Also - when I swapped to the N-SMA adapter, rather than the jumper, I noticed my noise changed (closer to 0 is better?) But that’s why I’m here, if someone could help me interpret what I’m seeing - that’d be great.

Here’s my weekly graph - ignore 05/06 - 08/09 is when I removed the jumper and went to the adapter

Here’s my hourly - I just swapped back to the jumper and removed the adapter to test the change in noise (2030).

The best way to tell is simply to experiment. I personally would try to increase that gain value up another 3 dB. In my experience, I think you have some room left. If you see improvement, then might try another 3 dB. Give it a day or two with any change.

Regarding the noise, you wish it to be negative infinity but that’s not possible and not the driver when tuning. The important thing is how high the signal is above the noise number, signal to noise (S/N). The signal can’t be too high though otherwise it swamps the receiver.

So I just did that I I find mine is at 0. I guess that is ok

Statistics: Tue Apr 12 05:04:23 2016 UTC - Tue Apr 12 06:04:23 2016 UTC
Local receiver:
8640528384 samples processed
0 samples dropped
0 Mode A/C messages received
267279 Mode-S message preambles received
56896 with bad message format or invalid CRC
125905 with unrecognized ICAO address
81155 accepted with correct CRC
3323 accepted with 1-bit error repaired
-21.9 dBFS mean signal power
-5.9 dBFS peak signal power
0 messages with signal power above -3dBFS
Messages from network clients:
0 Mode A/C messages received
15297 Mode S messages received
0 with bad message format or invalid CRC
0 with unrecognized ICAO address
15297 accepted with correct CRC
0 accepted with 1-bit error repaired
99775 total usable messages
0 surface position messages received
18593 airborne position messages received
18108 global CPR attempts with valid positions
0 global CPR attempts with bad data
0 global CPR attempts that failed the range check
0 global CPR attempts that failed the speed check
28 global CPR attempts with insufficient data
360 local CPR attempts with valid positions
360 aircraft-relative positions
0 receiver-relative positions
125 local CPR attempts that did not produce useful positions
0 local CPR attempts that failed the range check
10 local CPR attempts that failed the speed check
0 CPR messages that look like transponder failures filtered
14414 non-ES altitude messages from ES-equipped aircraft ignored
58 unique aircraft tracks
7 aircraft tracks where only one message was seen
CPU load: 20.1%
231650 ms for demodulation
453468 ms for reading from USB
38039 ms for network input and background tasks

It means it is too low.

Your levels seem very low. Do you have a long coax run? Are you in a remote location?

-21.9 dBFS mean signal power
-5.9 dBFS peak signal power
0 messages with signal power above -3dBFS

If you can increase your gain then I would try to add a few db. You will see more aircraft.

I have since then. Finding the sweet spot is tough, especially since my MLAT numbers are somewhat meaningless now.


First of all for the lazy ones: here’s a bash command that extracts the 2 relevant values from dump’s logfile and stores them into a csv:

egrep "(messages with signal power above -3dBFS|total usable messages)" /var/log/dump1090-mutability.log | awk '{print $1}' | sed 'N;s/
/;/g' > /tmp/dump_log.csv

To topic:
I’m using the Pro stick but have no success in getting any better results than before.
My Setup:

Cantenna (on roof) - InLineAMP 20db - coax (12 meters) - fa-filter - fa-pro-stick - pi

Tried lot’s of different gain settings from -10 via 1.7 up to 44.5.
With none of them I’m getting better reception than before with a nooelec usb stick :-/

Any ideas? Gainis set to -10 at the moment.

Here are some charts. ProStick has been installed last saturday.



You are not likely to improve on a setup with a mast amp with a low end USB dongle and filter.

Maybe the airspy or hackrf, Beast or flightfeeder could do better. Maybe a well tuned antenna.

The prostick with a filter and simple antenna is a simple, inexpensive and cost effective solution that should satisfy 90+% of users. Put the antenna up high and it will do wonders.

@jonhawkes2030: thanks!

Next planned step to improve is awtting my coco-antenna on the roof instead of the can.

Long run due to a lot of limitations. Gain max. That’s why I am anxiously awaiting the new FA amped stick.

What kind of cabling are you running? Even with only a 20’ I saw a huge improvement going from CA200 to CA400. You might consider upgrading it.

That said, CA400 is not cheap, nor is it particularly small or easy to work with.

Im using a High Grade RG6 (satellite grade) and its 25ft long. That being said in the KPAE region I run either 1st or 2nd in Number of A/C seen and in the top 10 in position reports. So over all Im pretty happy. Im looking forward to the FA Pro Stick … Just waiting for Amazon to get them in stock to order it.