Flarm suppression of ADS-B out on Tow Planes

On Friday, March 22, 2019 at 3:07:18 PM UTC-5, Darryl Ramm wrote:
On Friday, March 22, 2019 at 11:04:34 AM UTC-7, kinsell wrote:
On 3/22/19 1:44 AM, Ramy wrote:
This sounds like an interesting possibility as the powerflarm core has a USB port. I’ll let Darryl analyze it...

Ramy


The PingUSB just uses USB for power, data goes out over Wi-Fi, typically
to a tablet.

Right. And a third party MUX box that can receive that WiFi could merge it with FLARM serial data. But ah if the data is lower-resolution ADS-R or TIS-B targets does that MUX take it upon itself to generate FLARM $PFLAU target messages? Can it even tell what the source of the data is? Does it assign higher $PFLAU priority that FLARM targets? Does it rewrite existing FLARM generated $PFLAU messages? How does the behavior of those algorithms compare to FLARM's proprietary implementation? Does it only generate $PFLAA messages for this other traffic? That may be OK but what happens when a target gets close and the user does not get the high-priority FLARM warning they might otherwise expect? How will different FLARM displays handle any strangeness here?

Again it depends on what you are trying to achieve. The GDL-90 protocol serial data provided by most of these ADS-B out devices does not contain all the information you would want to build the most comprehensive integrated FLARM and ADS-B In system. You can provide coarse traffic information akin to what ADS-B In systems do today, and that may be fine but there are critters hiding below the surface of the swamp even there. It's possible to provide better integration if working on the "other side" of the interface, e.g. before the information is tuned into GDL-90 serial data, and that's a benefit of say the approach taken with the FLARM provided FLARM chipset and 1090ES receiver chip. That unfortunately does not get you UAT-direct. I just can't imagine there is a way for FLARM to invest time in UAT hardware for the small USA glider community.

I am glad some progress is being made but just want to caution not all this stuff is a simple as it might seem (Hi Andy :-)) . If it was we'd likely have stuff doing this today.

I was going to chime in since we built some Statux boxes (with a Ping USB receiver in my case which communicates with Stratux through the RPi USB port). All the problems Darryl describes are true. There is no way to de-duplicate and, depending on your setup and the traffic around you, a single UAT target may show up as 1090ES via ADS-R, TIS-B or UAT direct - all within a few seconds. Similar story for 1909ES targets. Other traffic with the other ADS-B Out physical layer can light up duplicate target for you. In addition, you have to roll your own alarms, which will be different from Flarm, because ADS-B delivers different data than Flarm so you may be stuck with simple proximity alarms (good luck writing your own mux software to de-duplicate Flarm native versus Flarm ADS-B versus Stratux/Ping ADS-R via UAT versus Stratux/Ping TIS-B 1090ES for the same aircraft (I demoed this mess for myself more than once) and God help you if that aircraft is your towplane. Also TIS-B resolution is only a mile or so depending on speed and distance from an SSR ground station plus network lag so you need a way to display it that shows you this imprecision.

At the time I spoke the Flarm guys about plugging PingUSB into the Flarm USB port since that's how it transmits data and would give you at least a fighting chance at integrating all the traffic types. It seemed like there were a lot of obstacles and I expect not the least of it is that the US only represents a few thousand units and it's a lot of custom code and may overwhelm the Flarm processor. I'm thankful they rolled a new OEM chipset that has the potential to support ADS-R and TIS-B. As long as you have ADS-B 1090ES Out that solves 99.9% of the scenarios anyone could care about. I wouldn't hold my breath for anything more. There's just no reason for them to accommodate UAT die-hards if ADS-R and TIS-B are more practical routes to getting the same job done.

Andy Blackburn
9B