►
From YouTube: ORI FPGA Standup 21 June 2022
Description
Breakthroughs all around. Plenty of good news and presentations by the team.
A
C
B
B
This
uses
the
opus
coder
instead
of
the
codec
2
coder
for
higher
voice
quality,
but
it
means
higher
bit
rates
and
all
the
numerology
has
to
change.
So
we've
done
that
to
the
modulator
at
least.
We
think
so
we
haven't
been
able
to
test
it
yet
because
we're
still
working
on
doing
it
for
the
demodulator,
it
turned
out
to
be
harder
to
understand
the
code.
So
we
went
off
on
a
side
quest
to
document.
B
B
B
Yes,
that's
also
tangentially
fpga
related
we've
been
talking
for
years
now
about
how
users
to
this
digital
transponder
can
be
filtered
by
the
by
policy
so
that
only
authorized
users
are
allowed
in.
Usually
that
means
everybody's
allowed
in,
but
maybe
a
few
bad
actors
are
are
excluded,
but
in
some
cases
it
may
mean
only
a
certain
set
of
people.
B
Maybe
people
involved
in
an
emergency
communication
situation
aren't
allowed
to
use
a
transponder
in
any
case,
by
being
fully
digital,
we
have
the
opportunity
to
do
more
of
that
than
it
has
ever
been
done
before
an
amateur
radio
satellite
and
we're
going
to
try
to
take
advantage
of
that,
and
one
of
the
key
insights
we
had
early
on
was
that
we
need
a
public
key
infrastructure,
in
other
words,
a
way
of
getting
trust
that
we
know
who
is
who-
and
it
turns
out.
That's
already
been
done
by
the
awrl
in
the
logbook
of
the
world.
B
They
have
issued
an
rsa
certificate
to
everybody
who
cares
to
have
one
in
the
m
radio
community
and
will
continue
to
do
so
for
qsl
purposes,
but
that
puts
the
crypto
in
people's
hands.
They
have.
They
have
a
certificate
or
they
can
get
one.
That's
actually
been
validated
by
real
world
procedures
in
the
united
states.
That
involves
mailing
a
postcard
with
a
password
on
it.
So
you
know
the
person
who
controls
that
postal
address
controls
that
certificate
anyway.
B
B
Now.
I
have
there's
a
jupiter
notebook,
which
I
checked
in
last
night
to
the
repo
that
uses
openssl
command
line
primitives
to
to
do
exactly
that
and
sure
enough,
it's
possible.
It
doesn't
take
any
special
cooperation
from
awrl
or
logbook
of
the
world.
The
certificates
you
need
to
do
that
are
are
out
there
in
public
on
purpose
and
not
just
leaked
or
anything,
and
I
showed
how
to
do
the
steps
that
are
going
to
be
necessary.
B
A
Forward
yeah,
thank
you,
we'll
we're
planning
to
have
a
post
recession,
presentation
at
the
rf
village
at
defcon
in
august,
and
hopefully,
since
this
village,
our
village
is,
is
dedicated
to
security
for
radio
frequency
work,
and
this
is
a
security
related
issue.
We're
hoping
to
get
a
lot
of
eyeballs
on
it
and
a
good,
solid
review
from
the
community.
A
And
you
know,
even
if
we,
if
we
don't
have
anything
further
than
what
we
have
today.
I
think
the
the
jupiter
notebook
and
a
poster
will
be
a
good,
a
good
thing
to
put
in
front,
but
we
still
have
quite
a
bit
of
time
so
hoping
to
get
maybe
a
more
of
a
demonstration
going.
So
it's
something
we're
looking
forward
to-
and
you
can
see
this
in
the
the
opulent
voice
protocol,
that
there
is
a
part
of
the
protocol
where,
if
you're
challenged
for
authentication,
that
you
send
a
special
packet,
we're
thinking.
A
If
for
the
for
the
initial
uplink
protocol,
we're
just
looking
at
putting
opus
frames
at
the
physical
layer.
But
you
can
see
from
the
document
that
there's
multiple
layers
and
what
that
means
is,
since
there's
there's
a
udp
layer,
ethernet,
ethernet,
udp
and
then
rtc
or
rtp.
Sorry
that
the
the
replies
for
the
authentication
challenge
would
go
to
a
particular
port
or
address,
and
so
that
reduces
the
need
for
for
using
up
bits
in
the
physical
layer
header.
A
So
all
of
this
stuff
feeds
into
the
implementation
for
the
for
the
fpga
at
the
in
the
payload,
and
all
this
is
active
discussion.
So,
if
you're
interested,
if
you're
listening
to
this
and
interested,
then
join
us
on
slack
or
join
the
mailing
list
and
and
we'll
be
regularly
posting
updates
pretty
much
every
week.
B
A
B
Mean
one
more
to
make
about
that,
the
primitives
are
are
well
known,
well-vetted,
it's
what
everybody
uses
on
the
web,
but
open
ssl
is
where
the
crypto
comes
from.
So
that's
not
in
question.
The
question
is
the
protocols
and
procedures
it's
very
easy
to
make
us
accidentally
put
a
back
door
or
make
a
screw
up
in
the
protocols,
and
so
we're
hoping
to
get
enough
detail
into
onto
our
poster
that
that
the
experienced
security
people
would
definitely
be
able
to
look
at
and
say.
Oh,
you
did
this
regular,
this
very
common
mistake.
B
D
E
Nothing
much
to
report
this
week.
I've
been
involved
in
some
documentation,
work
for
applying
for
grants
and
expression
of
interest.
We'll
continue
to
do
so
next
week.
Apart
from
that
had
a
discussion
with
leonard,
he
has
suggested
some
way
that
we
can
write
python
script
and
interact
with
io
drivers.
E
E
A
Yeah,
thank
you
so
much
so
yeah
I'll
give
a
brief
report
about
the
integration
effort.
What
onshore
has
set
up
is
a
repository
that
pulls
in
the
analog
devices
reference
design
for
the
zc,
706
and
adrv
9371,
and
also
stitches
in
the
encoder
from
swato
and
others,
and
this
hadn't
been
working
for
me,
and
so
what
we
did
over
the
past,
essentially
over
the
past
weekend,
was
to
really
dig
in
and
and
found
out
that
there
were
some
some
missing
files
and
that
the
other
some
hard-coded
paths.
A
This
is
the
same
encoder
that
everest
recently
demonstrated
at
friedrichshafen
for
over
the
on
the
pluto
that
works
on
the
pluto.
That
also
uses
a
reference
design
from
from
analog
devices.
It
turns
out
that
that
leveraging
the
reference
design
has
has
been
good
for
us,
some
other
information
about
the
design.
This
is
the
encoder,
which
is
the
transmitter
side
on
the
downlink
for
the
payload.
A
R
d
firm
was
interested
in
making
this
work
better,
and
so
they
have
engaged
on
it
to
help
them,
develop
it
further
and
fix
some
things
that
were
a
little
bit
fragile
about
it.
The
good
news
is
that
this
is
all
going
to
be
done,
open
source
and
will
be
donated
back,
so
we'll
have
a
matching
decoder,
and
that
will
of
course
need
some
some
work
from
from
us
to
to
test
out
and
to
publicize,
but
that's
really
good
news,
so
it
brings
the
decoder
along
faster
than
I
thought
it
would
go.
A
A
So
if
you
need
a
low
latency
link,
for
example,
if
you're
you're
all
into
voice,
you
don't
really
care
about,
maybe
missed
packets
as
much
as
you
care
about
getting
things
delivered
on
time,
and
that
is
something
that
the
scheduler
is
going
to
handle.
So
this
ken
and
others
have
advised
that
this
is
this
really
needs
to
be
and
essentially
in
firmware.
A
A
A
So
over
the
next
week,
whatever
we
can
get
done
on
any
of
this
is
what
we'll
be
doing
and
lots
of
lots
of
things
going
on
and
tests
being
made.
I
think
the
the
impediment
of
being
able
to
feed
the
encoder
with
bb
frames.
That's
that's
getting
solved.
I
think
we're
getting
a
better
and
better
idea
of
how
to
do
that
in
on
the
processor
side
of
the
zinc
all
right,
that's
it
are
there,
let's
see
everest.
Are
you
able
to
to
talk
and
hear
us?
F
F
So
there
is
a
github
repository
right
now,
which
include
the
the
design
of
the
pluto
with
the
dbs2
in
color,
so
the
firmware
there.
There
is
some
binary
already
in
the
firmware
in
order
to
to
use
the
dbs
tune.
Color,
and
the
next
step
is
now
to
publish
the
source
code
of
that
in
order
to
to
have
a
better
way
of
contribution
and
in
order
to
adapt
it
to
the
the
other
project.
With
the.
F
E
F
Yeah,
yes,
so
right
now
there
is
some
binary,
so
it
is
hardcoded
for
the
I
o
for
the
pluto,
so
it
shouldn't.
I
don't
think
it
works
directly
for
another
platform.
That's
why
I
try
to
publish
it
and
then
you
can
just
modify
the
I
o
number
mainly,
and
so
the
binary
are
on
the
overlay
of
the
bead
root.