►
From YouTube: FPGA ORI Meetup 14 March 2023
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Hello,
everybody
and
welcome
to
the
open,
Research
Institute
fpga
standup
For,
the
14th
of
March
2023.
We
have
some
updates.
We
are
going
to
move
forward
with
purchasing
an
adrb
9000
to
9002
development
board
to
go
with
the
zcu
106.
This
will
expand
our
capabilities
for
fpga
and
RF
SOC
work
in
remote
labs
and
this
particular
chip.
The
9002
has
been
requested
from
a
couple
of
different
people.
The
particular
project
that
needs
it
to
be
purchased
is
Neptune,
we'll
be
talking
more
about
that
in
the
near
future.
A
It's
a
ofdm
based
project
that
is
in
the
Drone
category
and
has
an
open
protocol
that
will
be
published
shortly
and
some
some
interesting
challenges
in
terms
of
design
and
and
form
factor.
A
So
this
particular
board
the
adrv9002.
It
comes
in
two
variants,
so
it's
not
like
the
adrb
9371,
which
we
currently
have
working
with
the
zc
706
xilinx
baseboard
in
in
the
case
of
the
9002,
we
need
to
decide
between
a
300
megahertz
to
three
gigahertz
version
or
a
three
gigahertz
to
six
gigahertz
version.
Those
are
two
different
boards,
so
one
way
to
do
this
would
be
to
buy
one
version
for
one
lab
for
remote
South
and
buy
another
version
for
another
lab,
for
example,
remote,
West
or
any
other
lab.
A
That
would
like
to
be
established
that
has
a
fpga
board
to
attach
it
to
so
we'll
be
purchasing
this
as
soon
as
we
can
kind
of
figure
out
a
game
plan
on
on
the
frequencies.
I
have
asked
Leonard
who's
leading
Neptune,
which
one
he
prefers
and
we'll
at
least
get
that
so
in
terms
of
expanding
the
capabilities
of
our
remote
Labs.
That's
the
latest
news
on
that.
We
have
a
zcu
106
and
both
remote
lab
West
and
remote
lab
South
and
they're
they're
used,
but
not
not
used
to
their
fullest
potential.
A
I
would
say
so
this
will.
This
will
go
a
long
way
to
to
giving
us
capability,
and
some
some
published
work
with
the
ultrascale
Plus
variant
of
the
of
fpgas
from
Silence
are,
are
floating
license
for
Bravada
will
handle
this
no
problem
and
the
remaining
factors
are
people
in
time
and
and
projects
that
need
the
support.
A
So
that's
the
major
update
from
from
this
end.
There's
there's
other
updates
about
training.
We've
been
talking
about
putting
together
a
custom,
MathWorks
class
for
for
us
and
that's
moving
forward.
It
looks
like
we're
going
to
go
ahead
and
choose
the
remote
version
rather
than
try
to
get
it
to
be
an
in-person
class.
The
closest
in-person
site
for
for
us
looks
like
Los,
Angeles
and
I.
Think
it'd
be
a
pretty
tough
go
to
try
to
get
people
to
Los
Angeles
for
an
in-person
multi-day
class.
A
They
also
offer
them
remote.
So
we're
going
to
go
ahead
and
and
say
Yes
MathWorks.
We
would
like
for
an
SDR
fpga
HDL
coder
class,
to
train
us
up
to
fully
utilize
the
things
that
we
have
in
the
labs
and
this
this
covers
both
remote
lab,
South
and
remote
lab
West.
The
the
machinery
would
be
something
that
we
could
better
fully
use
so
we'll
go
ahead
and
and
press
forward
with
a
remote
class
and
and
try
to
schedule
that
as
soon
as
possible.
A
All
right,
that's
the
the
big
updates
there.
I
don't
have
anything
else
in
any
other
in
any
other
categories,
so
I,
but
I
do
know
that
we
have
a
lot
of
Uplink
work
going
on
for
the
transponder,
so
I'll
I'll
turn
it
over
to
Paul.
To
talk
about
that.
B
Okay,
before
I
talk
about
that,
I
have
a
couple
of
follow-up
items
on
the
9002
board.
If
that's
okay,.
B
B
It
has
a
list
of
compatible
eval
platforms
of
two
and
that
one's
not
one
of
them.
So
how
confident
are
we
that
that
we
can
make
it
work
on
that
board?
Yeah.
A
So
when
you
navigate
through
and
and
if
you,
if
we
wanted
to
set
it
up
the
same
way
that
we
did
with
the
zc
706
and
the
9371,
we
would
navigate
through
the
the
HDL
reference
design
and
then
compile
that
particular
reference
design.
A
However,
we
can
also
support
this
directly
through
through
Matlab
and
simulink,
and
develop
HDL
code
that
way,
so
the
combination
of
the
zcu
106
and
the
adrb
9002
appears
and
and
and
is
a
viable
support
option
for
for
development,
and
it's
the
the
9002
was
specifically
identified
by
Sasha
and
Leonard,
both
as
like.
Can
we
really
get
this
up
and
running
in
the
lab?
A
So
if
there
is
anything
that
we
need
to
do
to
to
kind
of
fill
in
the
gaps
for
the
for
the
zcu
100
series,
which
includes
the
102
104
106.,
so
we
have
the
106.,
then
then
I'm
game
to
to
do
it.
So
so
that's
that's
the
answer.
It
is
a
viable
combination.
It
looks
like
it's
supported.
There's
there's
answers
on
the
engineer,
Zone,
where
they
they
don't
say,
hey,
that's,
not
a
supported
combination.
They
just
simply
go
ahead
and
answer
the
the
problems
and
and
of
course,
like
any
other
development.
B
Okay,
another
thing
that
confuses
me
about
this
is
that
they
seem
to
change
their
naming
convention
before
the
adrv.
Prefix
meant
the
dev
board
and
it
looks
like
now:
it
means
the
device
and
they
use
an
eval
adrv
9000
to
description
for
the
for
the
dev
board,
with
the
connectors
and
stuff
on
it.
A
Right
as
long
as
we
get
a
board
and
not
apart,
then
I'll
be
happy.
Yeah
adrb
stands
for
Analog,
Devices,
radioverse
sort
of
like
universe,
but
radio,
and
it's
a
it's
a
marketing
term
that
has
morphed
over
time.
So
yeah
they're,
some
of
the
adrv
line,
appears
to
be
a
module,
just
the
module
itself,
and
then
some
of
it
appears
to
be
the
the
boards
that
we're
more
familiar
with,
which
are
cards
that
plug
into
a
baseboard
with
the
supporting
fpga.
B
A
Yeah
it
should
I've
got
I've
got
a
couple
of
listings
from
like
from
from
digikey
and
and
Mauser
that
I've
been
looking
at
so
we'll
we'll
narrow
it
down.
We're
not
gonna
just
order
anything
willy-nilly,
but
we're
gonna
move
forward
with
this
particular
family.
A
There's
the
9002
there's
a
9004
or
sorry,
nine
thousand
two
and
three
and
four-
and
it
looks
like
those
variants-
are
just
like
the
numbers
of
transmitters
and
receivers
to
me,
but
we'll
we'll
we'll
review
the
the
actual
order
pretty
carefully,
but
we're
gonna
go
ahead
and
move
to
the
next
generation
of
of
radio
chips
for
the
for
the
Drone
project.
B
Okay,
oh
I,
I
see
the
picture
of
the
back,
makes
it
clear.
It
does
have
that
connector
I
was
expecting
it
to
be
a
through-hole
connector
because
it
has
so
many
I
think.
Well
anyway,
I
was
and
I
was
wrong.
It's
a
surface
mount
connector,
it's
mounted
on
the
back
of
the
board,
so
that
makes
sense
all
right.
Hopefully
somebody
will
be
around
to
tell
me
how
to
hook
this
thing
up
to
get
it
installed
in
this
remote
lab,
because
without
a
walk
through
from
Analog
Devices
I'll
be
over
my
head.
A
A
plug
it
into
the
correct
connector
power
it
up
and
we
have
a
radio
card
in
a
parallel
fashion
to
the
to
the
zc,
706
and
adrb
9371.
So
that's
what
I'm
expecting
to
see
and
that's
what
it
looks
like
it
will
be
to
me
and
if
anything
looks
weird
along
the
way
or
if
we
have
any
deal
breakers,
then
then
we
will
find
something
else,
we'll
we'll
go
we'll
we'll
move
to
an
another
radio
card.
A
So
the
basic
idea
here
is
that
the
the
Drone
drone
people
and
and
some
of
our
transponder
people
would
really
like
to
see
the
ultra
scale
Plus
be
in
the
mix
like
we
need
to
look
at
that
and
support
ultrascale
plus,
rather
than
stopping
with
the
7000
series,
which
is
on
the
zc706,
which
is
an
older
family.
A
It's
still
supported,
but
the
ultrascale
plus
is
pretty
darn
cool
and
that's
that's
where
some
interest
is
so
that
so
that's
kind
of
the
context
of
why
we
would
want
to
go
ahead
and
and
get
a
radio
card
for
the
for
the
ultra
scale.
Plus
board.
A
Yeah
yeah
I
don't
want
us
to
bite
off
anything
that
we
can't
chew,
or
that
isn't
something
that
that
we
can
support
I'm
willing
to
do
like
some
glue
logic
and
some
some
some
extra
extra
work
or,
if
there's
some,
some
files
that
have
to
be
written.
That's
fine,
you
know,
but
but
anything
way
out
of
the
ballpark
I
don't
want
to
do
and
so
far
this
looks
like
it's
it's
well
within
our
abilities
to
to
support
and
would
give
us
some
some
greatly
expanded
capabilities.
B
B
It's
not
connected
to
the
Spectrum
analyzer
at
the
moment,
because
the
Spectrum
analyzer
has
been
in
doing
useful
Duty
on
other
projects,
but
it
can
be
connected
to
this
Vector
generalizer
on
request.
If
anybody
is
ready
to
to
do
that
kind
of
a
test.
So
just
let
me
know,
and
on
the
subject
of
Uplink
work,
there
has
been
some
making
incremental
progress
toward
having
a
an
upline
capability.
The.
B
The
primary
focus
at
the
moment
of
active
development
is
on
incorporating
the
protocol
layers,
because
the
protocol
layers
will
have
a
fairly
significant
impact
on
the
numerology.
B
This
is
going
a
little
slower
than
it
might
and
slowing
it
down
with
some
some
head
scratching,
but
it'll
make
progress
and
we'll
have
that
working
for
too
much
longer.
I
hope.
B
Changing
the
frame
rate
also
involves
changing
much
of
other
things
like
changing
the
inner
labor
changing
the
the
randomizer
chain,
maybe
even
changing
the
the
coding
scheme
and
Michelle
has
been
very
helpful
in
in
giving
me
new
parameters
and
new
new
designs
for
those
things
and
there's
still
work
left
to
be
done
in
that
area.
A
No,
that's
all
good
we're
hoping
to
get
this
design
into
into
the
fpga
OR
into
fpga
code
as
well,
and
and
that
that
leads
to
our
upcoming,
like
classes
and
and
training
for
math
Works,
in
order
to
Leverage
The
the
tools
that
we
have
available
to
us
through
through
Matlab
and
simulink.
So
we
we
started
on
a
a
model
for
the
Uplink,
that's
in
in
Matlab
and
and
simulink
and
in
sort
of
in
concert
and
in
combination
with
the
the
modulator
and
demodulator
code.
A
A
You
know,
and
so
all
of
that
kind
of
kind
of
relies
pretty
heavily
on
some
some
very
interesting
tools
and
and
and
toolboxes
from
from
Matlab.
So
it'll
be
more
about
that
pretty
soon
so
far
so
good
we've.
We
have
some
outstanding
questions,
questions
that
that
we
don't
have
the
answers
to
yet
and
there's
there's
just
a
lot
of
a
lot
of
really
neat
work.
A
Going
on
to
try
to
answer
them
and
to
to
narrow
down
some
of
these
specifications-
and
you
know
things
like
coefficients
and
and
filters
and
and
interleaver
polynomials
and
and
all
of
that,
so
it's
been
progressing
pretty
well.
B
Oh
and
one
other
thing
I
wanted
to
mention-
maybe
you
covered
this
before
I
got
my
audio
working
here,
the
question
of
whether
to
get
the
lower
frequency
or
higher
frequency
version
of
the
radio
yeah.
A
A
It
goes
I
think
from
300
megahertz
up
to
six
gigahertz,
so
the
9002
doesn't
it
covers
300
megahertz
up
to
three
gigahertz
and
then
there's
another
board.
You
buy
a
separate
one.
If
you
want
three
gig
to
six
gig.
B
Yeah
I'm,
sorry
to
decide
it
at
purchase
time.
My
inclination
is
that
the
lower
frequency
one
is
likely
to
be
more
versatile
for
the
kinds
of
things
we
need
unless
we
X
actually
need
to
be
emitting
RF
up
at
gigahertz,
just
because
we
there's
more
test
equipment
that
works
with
it.
There's
you
know
more.
B
A
Okay,
yeah,
that's
smart,
okay,
I've
got
that
written
down.
I'm
gonna,
I've,
I've
gotta
I've
got
a
email
out
to
to
both
Leonard
for
the
for
project
Neptune
and
for
out
to
Sasha
for
transponder
work
as
to
what
they
prefer.
So
since
these
Dev
boards
are
are
helping
us
at
essentially
at
base
band.
It
almost
doesn't
really
matter.
I
mean
it's
great.
A
To
have
microwave
capable,
like
the
three
gig
to
six
gig
is
good
and
it
would
it
would
let
us
do
things
that
are
nominal
Uplink
frequency
of
five
gigahertz,
but
most
of
this
work
that
we're
doing
is
base
band
anyway
and
and
having
300
megahertz
to
three
gigahertz
version,
wouldn't
be
a
disaster
so
I'll,
so
I'll
I'll
wait
until
I
hear
back
from
the
teams
that
want
to
use
this
gear
and
then
we'll
go
ahead
and
get
whatever
and
like
I
said
I
mean
we
could
get
one
of
each
and
put
one
at
each
lab
depending
on
what
the
labs
want
to
do.
A
You
know
that
would
be
if
we
had
more
money
than
than
time,
so
I
think
I.
Think
we'll
we'll
take
a
measured
approach
and
we'll
listen
to
what
everybody
says
and
and
I
think.
The
advice
that
that
we
go
for
the
lower
frequency
band
is
is
good
and
it
it.
It
does,
make
a
difference
to
be
able
for
people
to
be
able
to
work
with
the
frequencies.
A
Yeah,
okay:
well,
we'll
do
the
smartest
thing
that
we
possibly
can
with
the
with
the
expansion,
the
lab
expansion
and
and
and
get
this
done.
The
the
protocol
is
really
neat.
It's
called
Flex
link
and
the
physical
layer
protocol
document
will
be
released.
First,
it's
I
would
say
done
at
this
point.
The
data
link
layer
for
the
flex
link
is
not
not
really
done
yet.
A
There's
some
some
questions
on
going
from
the
physical
layer
to
data
link
layer
that
that
are
kind
of
specific
to
the
Drone
world,
and-
and
you
know
so
so,
my
recommendation
to
to
them
was
to
just
go
ahead
and
release
the
physical
layer
open
source
protocol
document
so
that
they
can
get
feedback
immediately
and
so
that
we
could
start
working
on
it.
A
So
it's
it's
definitely
ofdm.
It's
definitely
leveraged
pretty
hard
from
LTE.
It's
got
a
lot
of
a
lot
in
common
and
and
a
lot
of
it
will
look
familiar,
but
it
has
some
some
truly,
you
know
so
streamlined
and
elegant.
A
You
know
advancements.
The
the
primary
author
has
a
whole
lot
of
experience
in
digital
Communications
and
is
a
pro
works
at
Rhoden
Schwartz
and
has
written
a
couple
of
textbooks.
So
I'm
extremely
happy
that
we
have
something
like
this
in
our
community
and
that
we
can.
We
get
we're
going
to
get
the
chance
to
support
it
and
to
to
do
an
implementation
and
possibly
a
board.
A
So
it's
a
big
opportunity
for
us
and
so
whatever
the
right
board
to
get
in
the
short
term,
we'll
go
ahead
and
get
and
then
look
forward
to
delivering
a
really
nice
thing
that
will
help
Aerospace
and
drones.
A
A
Think,
though,
for
a
single
convolutional
code
for
our
Uplink
that
we're,
if,
as
long
as
we
pick
the
the
biggest
spread,
you
know
the
largest
minimum
interleaver
distance
for
that
particular
set
of
coefficients
that
we're
going
to
be
that'll,
be
the
ideal
thing
that
that
will
be
good
and
the
coefficients
that
we
have
are
are
literally
the
theoretical
maximum
for
the
for
that
particular
frame
length.
So
you
know,
as
long
as
as
that
is
maximizing.
A
That
particular
statistic
is,
is
all
that
we
need
to
do
since
we're
not
a
turbo
code,
we're
not
using
turbo
codes.
So
we
don't
need
to
really
worry
about
the
stack
up
of
randomization
and
I
mean
I'll,
keep
looking
at
it
just
to
make
sure
that
we're
not
doing
anything
silly
and
ask
around
you
know
and
get
a
good
review,
but
I
think
we're
set
in
terms
of
making
updates
for
the
uplink
protocol.
C
Thank
you,
Michelle
not
too
much
to
report,
especially
compared
to
the
Fantastic
reports
we've
seen
we've
received
in
regards
to
the
rest
of
the
meeting.
Today,
pardon
we've
been
completing
more
infrastructure
work.
We've
been
slightly
delayed
in
the
temporary
deployment
of
the
equipment,
as
we've
focused
more
of
our
efforts
on
the
outbuildings
out
here
at
the
remote
lab
South
compound
I
guess,
but
we're
continuing
more
work
and
where
things
have
sped
up
a
lot
in
regards
to
our
outbuildings.
C
A
A
Yeah
that'd
be
great
yeah.
Our
next
newsletter
will
be
April
1st.
So
if
you
have
any
any
news
or
photographs,
that'd
be
a
good
good
place
to
put
them
and-
and
we
really
really
love
April
Fool's
Day.
So
if
anybody
has
some
good
technical
Shenanigans
to
put
into
the
newsletter,
then
then
please
send
them
our
way
and
we'll
we'll
get
them
in
there.
A
Yeah,
thank
you
awesome
all
right
and
we
will
have
a
an
event
in,
for
essentially
remote
lab
South
will
be,
will
be
the
the
central
focus
and
leading
a
pretty
big
event
in
September,
so
I'm
mentioning
it
because
September
we'll
get
here
before
we
know
it
and
it's
the
IEEE
I
work
or
iwrc
conference.
A
This
will
be
in
Little,
Rock,
Arkansas
and
we'll
focus
on
chips,
act,
money
to
underserved
states
such
as
Arkansas,
we're
hoping
that
remote
Labs
South,
since
it's
in
outside
of
Little
Rock,
will
be
able
to
benefit,
and
at
the
very
least
we
would
like
to
contribute
to
promoting
the
open
source
and
research
and
development
work
that
actually
does
go
on
in
the
state
really
looking
forward
to
that,
we've
gotten
a
lot
of
the
we've
gotten
Logistics
taken
care
of
for
the
out-of-towners
coming
in,
so
we
have
a
place
to
stay
and
we
are
pretty
much
ready
to
to
start
working
with
the
organizers.
A
A
So
that's
that's
coming
up
that'll
be
after
after
Defcon.
Some
we've
done
some
work
this
over
the
past
week
to
to
make
our
Defcon
open
source
radio
showcase
happen.
A
So
everything
is
working
working
pretty
well
for
the
for
that
Big
Show,
you
know
so
no
snags
and
additional
information,
and
we
also
have
another
opportunity
to
show
off
our
work
at
the
international
microwave
Society
of
2023,
which
will
be
June
in
San,
Diego
and
so
there's
space
that
we
can
use
in
order
to
do
any
sort
of
demos
that
we
would
like
to
kind
of
spread.
A
The
word
for
for
open
source,
Digital,
radio
that
benefits
amateur
radio
along
with
a
lot
of
other
services,
so
been
spending
some
time
trying
to
to
get
all
that
working.
A
Okay,
that's
that's
pretty
much
all
I've
got.
Is
anybody
need
anything?
Is
there
any
I
know
that
remote
lab
South
needs
money
and
I've
got
that
on
the
docket?
For
for
our
board
of
directors
to
look
at
shouldn't,
be
it
shouldn't
be
any
trouble,
but
does
anybody
need
anything
outside
of
that
any
resources
or
stuff
or
or
help
or
roadblocks.
A
C
B
Is
a
little
excuse
me,
a
little
bit
of
Hardware
progress
on
using
the
Pluto
to
transmit
I
the
standard
demo,
the
hello
world
for
this
sort
of
a
radio
is
just
a
tone
or
two
tones,
or
something
like
that,
but
I
wanted
to
demonstrate
that
I
could
actually
feed
a
continuous
set
of
generated
samples
into
it.
So
I
came
up
with
this
slightly
more
sophisticated
hello
world
where
I'm
sweeping
a
town,
sinusoidally
they're,
red
Trace,
is
the
primary
signal.
Obviously,
and
the
lighter
colored
traces
are
are
spurious.
B
One
down
the
middle
is
the
local
oscillator
leakage
and
the
mirror
image
of
the
desired
signal
is
obviously
a
just
exactly
that
it's
down
about
35
DB,
which
I
think
matches
the
quantization
error
of
the
12
DB
or
12-bit
Dax
I'm,
not
sure
where
the
three
times
image
comes
from.
That's
the
the
much
fainter
sweep
that
goes
out
almost
to
the
edge
of
the
screen.
B
This
by
the
way,
is
a
real-time
feed
from
the
Spectrum
analyzer
in
the
remote
lab,
using
the
newer
way
of
viewing
the
Spectrum
analyzer
remotely
that
I
cobbled
together,
where
it's
videos
being
encoded
on
a
separate
Raspberry.
A
Yeah,
it's
cool-
maybe
maybe
we
should
have
this
as
a
maybe
on
the
website,
like
a
I,
know
that
it's
in
the
repo
like
this
instructions
on
how
can
you
log
into
our
Spectrum
analyzer,
like
I,
know
that
that
exists,
but
is
there?
Is
there
some
way
to
that
wouldn't
be,
wouldn't
overload
it
to
have
it
show
up
on
the
website
or
have
somebody
click
through
and
and
get
this
feed.
B
A
That's
pretty
neat!
That
also
reminds
me
that
that
we
also
have
something
else
working
in
remote
labs
and
that's
Maya.
A
Yeah,
have
we
done
anything
in
okay,
yeah,
but
that's
that's
a
pretty
powerful
bit
of
code
that
looks,
looks
good
and
fully
fully
utilizes
the
Pluto's
fpga.
It's
really
nice
to
see
projects
like
that
and,
as
we
found
out,
the
Maya
was
done
by
using
the
amaranth
framework,
which
is
a
way
to
go
in
a
python-centric
way
to
go
from
python
code
to
HDL
and
Dr.
A
Estep
has
used
this
to
program
Maya,
so
you
use
the
amaranth
framework,
and
so
I
I
asked
him
if
he
would
be
willing
to
be
the
a
guest
editorialist
for
the
April
newsletter,
and
he
said
yes,
so
so
he's
slated
to
write
an
article
about
how
to
use
amaranth
open
source
framework
and
how
how
it
went
sort
of
his
experiences
with
using
amaranth
to
produce
his
his
project,
Maya.
So
I'm
looking
forward
to
that
and
being
able
to
to
put
that
in
the
newsletter
and
spread
the
word.
B
A
Cool
okay,
well,
I
will
shut
down
the
recording
at
this
point
and
see
you
all
next
week
and
then,
if
anybody
would
like
to
stay
on
for
for
office
hours
for
a
bit,
I've
got
some
time
here
and
would
love
to
to
talk
more
with
everybody.