Comparing ADS-B Reception with the RTL-SDR, Airspy and Beast Receivers
Over on satsignal.eu the author has set up a page showing live statistics of his ADS-B reception for the RTL-SDR and Airspy software defined radios, and also for the Beast ADS-B receiver. The Airspy is a $199 software defined radio that many consider as a next stage up from the RTL-SDR, and the Beast is a ~$270 USD dedicated ADS-B receiver.
Unsurprisingly the results clearly show that the Airspy receives ADS-B signals significantly better than the RTL-SDR. However, what comes as a surprise is that it is actually appears to be outperforming the dedicated Beast receiver. In the tests with the outside vertical antenna, the Airspy running on a Raspberry Pi appears to receive a significant higher number of messages and also sees planes out to a further range.
Not too long ago the Airspy team released their ADS-B software for the Raspberry Pi 2. They write that this software uses the full 10 MHz bandwidth and can even decode messages that are overlapping one another. We’ve also been told by the Airspy team that the Airspy is already in professional use as an ADS-B receiver amongst several small airports.
In the future we hope to compare the Airspy against the RTL-SDR on ADS-B reception ourselves, and also compare it against the 8 MHz bandwidth SDRplay whose development team have also recently released a new ADS-B decoder, as well as the recently released FlightAware ADS-B Prostick RTL-SDR.
I did a similar test some years ago with a stacked ADS-B antenna, a filter and preamp combination and a splitter on the end, running an R820T dongle with RTL1090 and PlanePlotter and the dedicated RadarCape with PlanePlotter in parallel. The RadarCape seems to have the same RF front end as the Beast receiver I believe and surprisingly the dongle created higher framerates with that setup. When running it with an attenuator the RadarCape excelled.
Since the designer of the Beast and RadarCape is an expert in GHz technologies, who took several iterations in optimizing his design, it is hard to believe that the wideband dongle has the better sensitivity. So I think for some reason the Beast and Radarcape seems to be tweaked for maximum sensitivity and appears overloaded when used directly behind a preamp, while the R820T tuner totally needs it in order to reduce the noise and compensate its relative insensitivity.
I would love to see how these two compared units behave without the preamp and maybe even without the filter.
I believe the dedicated receiver will excel in that test.
On the other hand the newer R820T2 based RTL dongles are outperforming their predecessors, the R820T, by at least 10% framerate in my setup, so maybe even without the preamp it might really be very close to the dedicated ADS-B front end of the Beast and RadarCape?!
Many thanks for the test and best regards
Did anyone do any further tests on this?