Sunday, 13 November 2016

ADS-B: Pro Stick Plus vs LNA at antenna

Update Nov 13, 2016: added info to images due to comment.

Common wisdom is to place LNAs at the antenna.
Let's test a FlightAware's Pro Stick Plus against Premium dongles with LNA at the antenna, with long coax cable length.
Which setup gets more reports?

Test details

- Pro Stick Plus as the highest performance pure ADS-B receiver for $21, review here,
- blog v.3 and Nooelec SMArt as top premium dongles available now, had same ADS-B performance in a previous test,
- Janilab LNA with external power, gain set to 19 dB, for $25 incl. worldwide shipping,
- two 45 feet / 15m RG 59 Mil-STD coax into splitter, Nooelec pigtails and adapters,
- two Raspberry Pi 3 from a ModMyPi Starter Kit, so identical back-end computing, system variance less than 2%.
Antenna is a half-wave telescopic commonly available with dongle bundles, on a metal tin. Normally, I get around 20,000 position reports with this antenna and short coax lengths.
No user involvement, and no gain adjustment in PiAware.

Results v.3 used on Day 1, Nooelec SMArt on Day 2. Totals indicated.

Polar Plots:

Frankly, I'm just as surprised as you are, as I've only seen the results when double-checked which IP address belongs to which receiver combo.
15.17% percent more position reports with a Pro Stick Plus, which costs half as much as a premium dongle and preamp. Granted, it will be pretty much unusable for general use due to filtering, but the same 1090 MHz filter also does wonders for ADS-B reception.
Significantly less position reports than usual; other FA feeders in my area did not experience such a downturn, so the only explanation is long coax length. Minimize coax length for best signal, or completely eliminate it with direct connections.

It's only one test

Your location, preamp, antenna, cable and retina is different from mine. The "LNA at the antenna" mantra comes up so often that I wanted to see what happens for ADS-B use. That's all.
The fact is that a receiver built for a specific task with a preamp and filter, costing $21, beat a combo built for general use costing $46. Adding a filter for $20 might narrow the performance gap, but spending $66 when the same job can be done for $21 seems a bit stupid. If you can run coax to a remote location and mount an antenna there, you should be able to run Ethernet cable and mount Pi3 with PoE at the antenna. Ethernet cable is so much cheaper than coax.


  1. Just to be clear, you tested each setup on different days?

  2. No, I tested them on consecutive days, Sat and Sunday via splitter.Two day's data and Polar plots in the post. I'll update post to make this clearer, thanks!