Re: [Discuss-gnuradio] gr_correlate_access_code_bb question

[Robert McGwier]

Dan Halperin wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hey all,
>
> Sorry to bring up painful memories :). I did some calculations to work
> out frequency drift of the USRPs, and I wanted to know where I'm going
> wrong.
>
> The crystals in the RFX 2400 are spec'ed at 50ppm error, which is to say
> that the maximum frequency offset between the sender and receiver is
> then 2 * 50 ppm * 2.462 GHz (802.11b Channel 11) = 246.2 kHz.
>
> 246.2 kHz sampled at 64 MHz is 246.2 kHz * 2pi / 64 MHz which is
> approximately 0.02417 radians / sample.
>
> At, say, DBPSK 250kbps, oversampled by a factor of 2, that means we
> decimate by 128 (500 ksamples per second), to then get at most ~3.094
> radians per sample drift.
>
> With DBPSK this could flip all of our bits! Does that mean we in fact do
> need to worry about the case where the complement of the access code is
> detected? Is my understanding of the meaning of ppm error wrong? Does
> the distribution we assume for the frequency offset make this
> possibility negligible?

No it means you have to estimate the carrier frequency offset and remove 
it, even in DBPSK.  You do this by using the access code to find the 
frequency offset and the differentially detect it.


> Thanks!
>
> - -Dan

Bob


--
AMSAT Director and VP Engineering. Member: ARRL, AMSAT-DL,
TAPR, Packrats, NJQRP, QRP ARCI, QCWA, FRC. ARRL SDR WG Chair
"If you're going to be crazy, you have to get paid for it or
else you're going to be locked up." Hunter S. Thompson