Sorry, I misunderstood and though you were talking about. Now that I'm
not confusing what you meant I appreciate the thought process you are
going though but I want to try to emphasize that PCAS just cannot be
pushed very far to do really advanced things, certainly not with Mode
C transponders in larger gaggles. And Mode C likely starts having
problems with less than what many would consider a "large gaggle".
Things are a lot better if we have a Mode S environment but many
gliders in the USA are going to have Mode C transponders.  So I'll run
though the main issues below, kind of in a jumbled up way, hope it
makes sense....

In a gaggle type environment when an interrogation hits the gaggle all
Mode C transponders (an actual Mode C transponder or a Mode S being
interrogated as a Mode A/C) reply at the same time and their
transmissions overlap. This is called synchronous garbling and it only
takes a few overlapping replies for this to make the signal
unreadable. Forget what might be possible, the PCAS is quite likely to
just not be able to read the altitudes of any of the replies. SSR
radar and TCAS systems have fancy hardware decorrelators that help
them try to pull out signals from each other but they can only handle
a few close overlaps. SSR radar obviously also sweeps the
interrogation beam and TCAS II uses a  wider quadrant interrogation
(and other tricks) to interrogate as few transponders as possible at a
time. But any of these systems will just fail with reasonable large
number of Mode C equipped gliders in a gaggle.  Synchronous garbling
is the reason why there are procedures for formation flights with Mode
C transponders to have everybody but the leader squawk standby. There
has been research done on Mode C synchronous garbling of SSR radar in
glider gaggles by European researchers and the results were bad as
expected. Synchronous garbling is a really fundamental problems that
will severely limit PCAS usability against Mode C threats if you are
flying close to or in a gaggle type environment.

Gliders in a gaggle will all be squawking 1200 or 1201 not unique
squawk codes. Nobody flying those gliders is going to be on a flight
plan or under flight following and the FAA with limited Mode A
allocation blocks is unlikely ever to give away squawk codes for this
even if it would make any difference. And here is another reason you
don't want to do this (see below later).

It could be that most interrogations of a glider gaggle come from TCAS
systems. TCAS does not interrogate Mode A (although there is a slight
qualification on that I don't want to get into). It has no idea what
the squawk codes of aircraft are, all TCAS wants to know is a threat's
relative altitude and direction (which it gets from    phased array
antennas (those two blade next to each other you see on aircraft)). So
there are may be none or few Mode A replies to correlate Mode C reply
altitude replies with.

PCAS does do not actually listen on 1030MHz so never see the
transponder interrogations (just the replies on 1090MHz), it makes no
sense to bother to do this because the moment the replying transponder
is any distance away the PCAS would not know if the interrogation it
sees is what the transponder is actually replying to. So a PCAS unit
has no idea whether a Mode C transponder (an actual Mode C transponder
or a Mode S being interrogated as a Mode A/C) is replying to a Mode C
or Mode A interrogation. All it sees is a series of encoded pulses.
PCAS units know from the ranges of valid patterns that some reply
patterns must be Mode C replies and some must be Mode A. However there
is an significant overlap range where the PCAS cannot know for sure
what the reply actually is. PCAS units clearly try to do some tricks
to reduce this "aliasing" problem but vendors do not disclose how they
do this, an obvious filter is if the code does not change over a
reasonable time then it is more likely to be Mode A than Mode C. But
an auto-pilot with very accurate altitude hold may break that
assumption.  I'm pointing this out for completeness but it is actually
a non-issue for us since 1200 and 1201 (and 0440 for folks flying near
Reno, and all the emergency squawk codes etc.) do not alias to valid
altitude codes. It is a possible issue however with discrete allocated
squawk codes and I think I've seen it in practice with my Zaon MRX.
BTW this is the reason you would *not* want to try to have Mode C
equipped gliders in a gaggle with discrete squawk codes.

With Mode S transponders replying to Mode S interrogations things
quickly get more complex, but the good news is easy to correlate a
squawk code, and what is an altitude for a Mode S transponder replying
to a Mode S interrogation. But even better you don't need to worry,
you can mostly forget about the squawk code and instead use the
transponder's ICAO ID itself is the unique marker that differentiates
replies from different transponders.

Mode S transponders are a lot more advanced than Mode C transponders,
it's impressive that Trig can make the TT21 and get it to market at
what we used to pay for Mode C transponders (let alone that it also
does 1090ES data-out). There is no end-of life for Mode C in the USA
and Mode C transponders continue to give good visibility of gliders to
ATC and TCAS systems so I'd not hold my breath for lots of glider
owners to replace Mode C with Mode S transponders. Over time they may
do so to gain 1090ES data-out capability. But this stuff all starts
adding up in cost fairly quickly.

With a Mode C transponders (either an actual Mode C transponder or a
Mode S transponder being interrogated as a Mode A or CC) the is no way
in high density environments to really correlate an aircraft's
transponder Mode A interrogation reply (containing the squawk code)
and the same transponders separate Mode C interrogation reply
(containing altitude). The only thing that (unidirectionl) PCAS has to
go on is the strength of the RF signals and trying to guess that the
strength of a Mode A reply  matches a Mode C reply. If the PCAS unit
like the XRX has directional ability that can help as well, it might
be able to help but it is likely to get very confused when the threats
are moving around relative to the antenna. This problem crops up in
general in transponder in multiple other situations and is sometimes
called "code swapping".

There were was at least one early PCAS systems that did display threat
aircraft squawk codes and that was kind of handy (gave a clue if a
threat was on flight following/plan for example). They seemed to do
that that pretty well in typical GA environments, but that would fail
in dense gaggles for the reasons described above.

If interrogators are coming from a single approach radar then
interrogations will occur every 5 seconds. If from an area radar every
12 seconds. These radars will interleave multiple Mode A/C/S
interrogations across the gaggle in each rotation. But the PCAS even
if it can decode the replies (ie. is not killed by synchronous
garbling) will start having problems trying to "track" Mode C targets
by comparing RF power strengths  at these relatively large time
intervals given how sensitive the RF power received will be to
obscuration and relative antenna orientation. If it can decode the
replies it will have accurate altitude (+/- 100' for Mode C) for them.
And luckily in many places you'll get more frequent interrogation from
airborne interrogators or have overlapping SSR radar interrogations
(but most of those then won't be Mode A interrogations).

With Mode S transponders being interrogated by a Mode S system (most
radars and all TCAS systems, but there are still Mode C interrogators
out there) a PCAS system could use the Mode S ICAO ID to "keep track"
of separate aircraft. I have no idea if the PowerFLARM or other PCAS
do this. Once the Mode S transponders have done their initial
handshake with the interrogator they will be selectively interrogated
and so their replies will not suffer the same rapid synchronous
garbling problem that Mode C transponders do. Right now most
transponders in gliders are Mode C so synchronous garbling is likely a
very large issue. Also in very large gaggles like big contests you
will likely run into congestion problems with selective interrogation.
I'm not able to model that in my head and/or guess what the limit
really is. It depends on the mix of interrogations.

In a gaggle or dense glider type environment the received RF power of
another transponder depends on slant distance and also relative
orientation of both aircraft (especially with blanking effects) and
antennas. This is very different from PCAS working in a much more
stable type state environment with a few nearby powered aircraft
cruising along. I expect the distance guesstimates and filtering the
PCAS systems used to have lots of problems getting range accurately,
but then PCAS systems alarm mostly on altitude conflicts for this
reason.

(Getting off topic but..) For the glider on tow situation with two
Mode C transponders in a glider and tow plane you may get synchronous
garbling and see occasional strange jumps in the signals seen. Where
for for example at one point in time the PCAS might think it sees one
threat, (two Mode C transponders squawking the same altitude bit
pattern and mistaken for one transponder) then the as you climb a bit
later it sees a garbled reply (two transponders synchronous garbling
with slightly different bit patterns as one of the encoders rolls over
the next 100' mark). I suspect this is a cause of some of the
interestign things we see with Zaon MRX at times when on tow. And it
just varies a lot as to the relative strength of the towplane
transponder signal and the leakage of the local transponder seen by
the PCAS. Again with Mode S in both aircraft it would be relatively
trivial to disambiguate aircraft and deduplicate/suppress threats.

In a PowerFLARM environment with lots of gliders also with
transponders what you want to do is correlate the Flarm ID of a glider
with it's transponder signal (and also extend that to ADS-B as well)
and suppress the PCAS warning for that glider. This in principle is be
easy to do for Mode S. You just have have the Flarm transmitter also
transmit the aircraft's Mode S ICAO ID). With Mode C you are pretty
much out of luck. The Flarm box is only going to be able to guess at
best. And there is the trade off between suppression a duplicate alert
and suppressing a real alert. It is so insurmountable that personally
I'd not bother trying if I was developing the box. And to be clear, I
have no idea what Flarm does inside their PCAS system in PowerFLARM.

Hope that helps explain some of the issues.

Darryl