Performance decreased when antenna moved to the roof

FlightAware ADS-B Flight Tracking
1

(Correction: this post previously indicated I had a ProStick, it is actually a ProStick Plus (the blue one with the integrated filter)

For years, I’ve been dreaming of the day I would finally move my antenna to the roof and marvel at the increased range and performance. That day finally arrived—well, at least the first part did. The antenna made it to the roof, but the performance actually decreased.

I’m running PiAware with a ProStick Plus. The Pro Stick Plus is connected via a 50‑feet USB cable, and the FlightAware 1090 MHz antenna is connected to it through a 3‑foot coax cable.

I had this configuration running for years in my office next to a window. The long USB cable was coiled to avoid issues, the gain was set to auto (around 58), and I was getting a good mix of range (100–150 miles) and positions. I’m less than 20 miles from SeaTac, so I get plenty of nearby traffic.

To my surprise and disappointment, when I moved the antenna outside, my range not only failed to improve (despite now having a full 360° view), but overall performance actually dropped.

The graphs show that my noise level is now surpassing my weakest signal, which was very surprising—I expected the Pro Stick Plus to suppress most of the noise.

As you can see from the graph, the auto‑gain initially overwhelmed the signal, so I’ve been adjusting the gain to get back to the noise level I had before and about 10% strong signals. What I don’t understand is how my house was acting as a band‑pass filter. Any ideas what might be happening here? I haven’t gone back up to the roof to check the connections yet, but I doubt that’s the issue. I was hoping the Pro Stick Plus would do a good job filtering noise even outdoors. Really surprised and disappointed by this setback… years in the making.

By the way, there are no other antennas or devices on the roof, and the antenna is almost in the same spot as before—just 10 feet higher and outdoors.

image
image1360×710 252 KB

2

Well your USB cable did not reach all the way up? How is the antenna now connected to your dongle?

3

It is the same USB cable I was using before. It is 50 feet long! Amazon.com: LDKCOK USB 2.0 Type A Male to A Female Active Repeater Extension Cable 50ft, High Speed 480 Mbps : Electronics

4

Add a Flightaware Filter between coax and ProStick.

 

 

2 Likes
5

You may want to adjust your gain down somewhat. @abcd567 suggested a bandpass filter and I think that’s the best first step. Your gain, however, could be blasting your SDR, making good signals seem like noise. If signal/noise doesn’t improve after the filter, you might look at turning gain down a bit to see if that helps.

1 Like
6

Lower the gain to 48 dB, and install a FlightAware filter, preferably the dark blue version. Skip SAW filters for this job.

The house does not filter, but attenuates. Now the dongle is overwhelmed.

While the USB cable does not appear to be the problem, do you really need that long a cable?

1 Like
7

Thank you all for the recommendations. I was thinking a filter would be the next move, but I wanted to run that by everyone first. In my mind, I wasn’t expecting to need a filter since the ProStick Plus has one integrated.

I’ve been adjusting the gain values from the beginning; it was clear to me that the gain was overwhelming the system as soon as I saw the signal‑level graph. I started at 36.4 and have gone up and down—38.6 seems to be the best level so far (weakest signals between -31 and -25 dB, and around 10% strong messages).

As for the USB cable, I only need about 40 feet, but couldn’t find one of that length. It was either 30 feet (too short) or 50 feet, which is a little longer than necessary. I wanted to extend the USB cable and minimize the coax length to reduce RF loss. I don’t think the USB cable is affecting performance, since I was using the same cable before—it just wasn’t extended or routed through the walls and roof eaves. I’ll run some experiments this weekend to continue debugging and report back.

8

The ProStick filter is useful, but an external inline filter like the one above will give you better control over both program bandwidth and resonant frequency filtering.

1 Like
9

I would rather have a longer coaxial cable than a longer USB cable, but that is just me.

10

An external filter like the recommended dark blue one does the filtering prior to the builtin LNA and the filter.

For the Pro stick Plus the builtin internal filter is located after the builtin LNA (and that amplifies all the signals) and that is why your setup is overwhelmed.

By adding the external filter you create a clean stream of 1090MHZ messages that are then amplified and afterwards filtered again.

Current setup you have is antenna-cable-prostick plus(LNA-Filter)-Raspberry Pi

With the external filter you will create this chain:

antenna-cable-external-filter-prostick plus(LNA-Filter)-Raspberry Pi

That will give you a lot less overwheming signals and a better adsb signal going into the Prostick dongle.

1 Like
11

As long as the USB cable negotiates and runs at its intended speed it’s better to have a long USB cable because the kind of digital data being sent has no losses. I’d take a long USB cable over a long coaxial cable.

Unless you’re getting errors or reduced speed and that causes dropped samples.

2 Likes
12

Here is the update. I got the 1090 MHZ band filter from the FlightAware store as recommended by everyone in this thread. Since I also had a ProStick (Not the Plus) I also swapped the receivers. My logic is that if I am adding an inline filter, I don’t need another immediately in the receiver. I did this mostly because I am concerned with attenuating the signals too much and loosing range.

I am happy to report that my signal levels have improved significantly, and as consequence of that so has my range and positions.

Here is the graph with the signal and noise level. The orange lines delimit the moment the filter was installed. Notice how the noise level decreased from around -24 to around -35 db.

I have not changed the gain, but my give it a try after a couple more days of building a baseline.

image
image895×407 45.7 KB

3 Likes
13

Attenuating (within reason) after the LNA really isn’t an issue, using the blue stick should be fine.
The blue stick has the LNA first and the filter second.

But considering you have both SDRs this is a very simple A/B test if you’re at all curious.

2 Likes
14

My current chain of command: