Re: [Discuss-gnuradio] Latency measurement using GnuRadio

[Ed Criscuolo]

The short answer is, because our customers require it!

The longer answer is that these are strictly low-earth-orbit
spacecraft, communicating through a relay in geosync.  So,
the round-trip time is dominated by the round-trip time to
geosync, which is around 250 mS.  The new modems include all
sorts of fancy signal processing, including LDPC error correction
and direct-sequence spread spectrum.  So their processing time
could easily become significant.

The whole system has an overall latency budget, and a portion of
it is allocated to the modems.  They're being built out-of-house,
so we need to be able to verify their performance when we do
acceptance testing.

@(^.^)@  Ed


address@hidden wrote:
> OK, so I'm being a smartass, but why do you care about latency on a path whose 
> path length may be anywhere from a few 10s of milliseconds, to
>   several hours?
>
>
>
>
> ------- Original Message -------
> From    : Ed Criscuolo[mailto:address@hidden
> Sent    : 9/10/2009 1:38:07 PM
> To      : address@hidden
> Cc      : 
> Subject : RE: [Discuss-gnuradio] Latency measurement using GnuRadio
>
>  Hi all.
>
>    At work we're building some new modems for spacecraft communication,
> and I've been tasked with coming up with a test set or tool for
> measuring latency through each side of the modem (modulator and
> demodulator).
>
>    I'd like to use GnuRadio 2 and a USRP2 to do some or all of this.
>
> I was wondering if the sample time-stamping capability was fully 
> functional in version 2, and if I could use it to insure that a given
> sample was transmitted at a specific time.  Similarly, can I get time
> stamps on received samples?
>
> TIA
>
> @(^.^)@  Ed
>
>
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
>  http://lists.gnu.org/mailman/listinfo/discuss-gnuradio