[Ground-station] The Future of Military Radar: Panel report

Wed Jun 21 10:23:22 PDT 2023

There was a panel about the future of military radar today and we were
invited. Handouts attached:

Online Panel - The Future of Military Radar - Slides June 21.pdf

Analog Devices - Memorandum to the Working Engineers in the World.pdf

Rohde & Schwarz - An Introduction to Passive Radar Systems.pdf

Mercury - Strategies for Deploying Xilinx's Zynq UltraScale+ RFSoC.pdf

Notes:

Spectrum Sharing is a critical element here.
We’re on the Dynamic Spectrum Sharing working group at FCC TAC.
They agree.
As amateurs I think we know how military radar can wipe out a band.
our Q: What is the role of reflectarrays in the future of radar? We
generally believe that if you can afford to use a phased array, you
probably should, but reflectarrays seem really useful when you can’t afford
the complexity or power consumption of phased arrays. In the context of the
future of radar, is there a path forward for reflectarray implementations?
A: Many things will exist in the future. There’s a wide open architecture
for future radar, and many ways things can be done. Depending on all the
many complex requirements, there will be opportunities for way the things
will be done will include the way it was done in the past, and the way it
can be done in the future. Analog/Digital boundary, simple/complex,
rotating elements/fixed elements. All of it can exist, and it’s application
specific. “It depends”
A: reflectarrays are a interesting tech that’s been around a while and they
address some of the issues of parabolics but still need some steering. A
trend? potentially using some of the metamaterials in a reflectarray to
steer the beam to change the properties of the reflector itself. Couple of
discussions over the years, but they don’t yet address the SWAP of the
larger phased array antennas.
Main use has been on satellites.
Physical space is such an issue. Therefore it’s a great answer.
“How do I reduce SWAP for each of the elements of our radar system” <= an
increasingly important metric for radar.
Reiterated again.
Customers are moving from simple to complex modulation schemes, and
including targeting information into the radar itself. This means vector to
scalar transitions. Huge challenge.
And additional burdens to dynamic spectrum allocation.
“Is the data what I expected it to be” <= in the system budgets for this
stuff, when you are trying to get info out of the air, with all this RF
frontend stuff and digital processing, there’s a real fear that what pops
out the other end is nonsense or not.
being able to visualize it? Is not where it should be.
There’s at least as much processing required for the calibration, than the
actual observation.
These people must consume a lot of blood pressure medication.
“Things get miniaturized… it doesn’t work… the first thing you do is take
it apart. When you can’t take it apart, what do you do?”
This is where the approaches from the ASIC crowd become relevant, I think.
“Precompiled complex waveform patterns broadcast in realtime” was a
sentence uttered in the context of battlefield readiness.
Q: DARPA has a program BLiP (Beyond linear processing) which is about
cheaper processing power. How relevant is this?
A: anything that reduces processing requirements is great (kind of an end
run around a very general question)
Q: how is GAN tech affecting system and test requirements?
A: GAN relative to the block diagram of these systems: why the trend?
Higher power density. That’s big. Higher voltages, more efficient, big
trend towards it.
Some of the limitations? Recovery time, particularly on the LNAs. GAN LNAs
have higher “survive power” but the recovery from high power blasts is kind
of a deal.
at the moment, this is the tradeoff and consideration.
Cost and size issues too.
SWAP (c) with the c being cost is a concern.
Yes you get SWAP but the cost is still painfully high.
8:45
How does the trend towards GAN affect test equipment? GANs aren’t the be
all end all, but they are significant part of the landscape.
Performance varies on thermal conditions. heat distorts phase.
you have to deal with this in order to get the most out of the tech. DPD is
affected by this, for instance.
DPD = digital pre-distortion
which is super important for higher performance digital communications.
comparing between multiple pulses, since the material itself changes with
respect to thermal conditions, you have to adapt your radar (or comms
system) to take these physics into account.
otherwise you are leaving a lot of gain/performance on the table.
it looks great at the brochure, but when you pound on your transmitter, it
droops and changes and squirts away from you.
you can end up with crap performance if you don’t move with the molecules.
(I didn’t know this)
no such thing as one size fits all in the analog signal chain for modern
radar systems.
this is why there’s such a dizzying array of options in the “market”
They are now talking about quantum radar.
“not ready yet”
Q: can you do modern radar with current FPGAs?
A: What could a radar look like without an FPGA? Not possible.
Maybe automotive radars, where everything is an ASIC, is the only example.
ADI + Xilinx is the combo.
Multiple different processing engines can be chosen to address different
tasks on the fly. This heterogenous approach only exists with FPGAs+ADI
chips (at this time)
FPGAs aren’t going away anytime soon (for military radar)
they are just going to get more and more capable and sophisticated and
larger more high bandwidth memory, more AI/ML functions - because this is
exactly what functions are required by the biggest customers. Which are, at
this time, military industrial.
Pushing things closer to the antenna is the goal, and this is happening.
(edited)
Synchronization of all the processing - if it can be done - go ahead and do
it. FPGAs are made for a lot of parallel streaming logic.
digital blocks get more and more defined - they are included right behind
the converters.
Now that’s it’s well defined - as soon as that happens, then you commit to
silicon.
If you need programmability (and this is a field that does) then FPGAs are
still absolutely necessary.
In the commercial world, there are balances between FPGAs and ASICs, and
“we” (military world) needs to watch this.
8:58
There’s always some level of programmability (adaptive) that people need to
have. That’s the metric.
Configurability of the logic is a unique aspect of FPGAs.
Something interesting to me - MIMO is largely written out. It’s considered
to be “a failure”..
Q: What are you doing from defense department contacts what the viability
of military radar based around high value targets when we have a completely
different landscape with “loitering” opportunistic targets?
A: high value fixed assets vs. mobile opportunistic ones.
A: right?
Q: yes
A: I’ll take a stab at it. I’d talk to people building these systems that
are built for prime contractors and deal with the questions of “how many
aircraft carriers do you need” vs. “distributed swiftboats”
A: even when you have a big array, it’s still represented by distributed
receivers. When things are distributed and flying around like drones, you
can’t really do coherent stuff. But we’re not in a world where we can
define that. We have to simply provide what we can provide and adapt. The
world has changed.
(it’s a messy tactical world and big phased arrays on aircraft carriers may
not do the job any more)
an 80% solution is now a huge win vs. the 90% solution with big iron money.
Finally Ukraine mentioned
Drones and drone tech, cheap and fast, “gets the job done”
small platforms with “good enough” DOF DOA Battle Assessment etc.
“where are the emitters in the environment and how can we find them”
“these are things giving us pause and requiring us to consider how to deal
with future operations” (no kidding, Sherlock)
Someone asked a question that resulted in a “can’t go into that into an
open session” - with respect to calibration of radar systems over the air.
This is stuff usually done only in an anechoic chamber, but my read on this
is that it’s a live function that this particular vendor can do.
Repeatability over long periods of time, and OTA calibrations of phased
array radars, are a topic here. One of the panelists is going to publish a
white paper about this in the near future.
Calibration is a big factor, as mentioned several times.

Ok well that was a lot!

Some of it is relevant to how we have to share our bands as amateur radio
operators and experimenters.

Comment and critique welcome and encouraged.

-Michelle Thompson
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openresearch.institute/pipermail/ground-station-openresearch.institute/attachments/20230621/70cca82f/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Analog Devices - Memorandum to the Working Engineers in the World.pdf
Type: application/pdf
Size: 179963 bytes
Desc: not available
URL: <http://lists.openresearch.institute/pipermail/ground-station-openresearch.institute/attachments/20230621/70cca82f/attachment-0004.pdf>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Online Panel - The Future of Military Radar - Slides June 21.pdf
Type: application/pdf
Size: 618104 bytes
Desc: not available
URL: <http://lists.openresearch.institute/pipermail/ground-station-openresearch.institute/attachments/20230621/70cca82f/attachment-0005.pdf>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Mercury - Strategies for Deploying Xilinx's Zynq UltraScale+ RFSoC.pdf
Type: application/pdf
Size: 717404 bytes
Desc: not available
URL: <http://lists.openresearch.institute/pipermail/ground-station-openresearch.institute/attachments/20230621/70cca82f/attachment-0006.pdf>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Rohde & Schwarz - An Introduction to Passive Radar Systems.pdf
Type: application/pdf
Size: 3874660 bytes
Desc: not available
URL: <http://lists.openresearch.institute/pipermail/ground-station-openresearch.institute/attachments/20230621/70cca82f/attachment-0007.pdf>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Screen Shot 2023-06-21 at 08.10.25.png
Type: image/png
Size: 144105 bytes
Desc: not available
URL: <http://lists.openresearch.institute/pipermail/ground-station-openresearch.institute/attachments/20230621/70cca82f/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Screen Shot 2023-06-21 at 08.27.03.png
Type: image/png
Size: 1527740 bytes
Desc: not available
URL: <http://lists.openresearch.institute/pipermail/ground-station-openresearch.institute/attachments/20230621/70cca82f/attachment-0003.png>