Hello,

I am using blks2impl dqpsk modulator/demodulator.

Since demodulator’s “samples_per_symbol” value must be an integer >=2

and

USRP samples at 64 Msamples/s with a 2^(n) possible decimation, I have

to

sample a 2100 kbps carrier (=1050 ksym/s).

I found two blocks that may apparently be used for resampling:

- rational_resampler()
- rational_resampler_base_xxx()

I am not sure of which one I should use. Does somebody know the

difference

between them?

I’ve read that rational_resampler_xxx() is a high-level wrapper for

gr.rational_resampler_base_xxx but I don’t understand what it means.

–

View this message in context:

http://www.nabble.com/Resampling-tp17643534p17643534.html

Sent from the GnuRadio mailing list archive at Nabble.com.

On Wed, Jun 04, 2008 at 03:21:25AM -0700, irene159 wrote:

Hello,

I am using blks2impl dqpsk modulator/demodulator.

Since demodulator’s “samples_per_symbol” value must be an integer >=2 and

USRP samples at 64 Msamples/s with a 2^(n) possible decimation, I have to

sample a 2100 kbps carrier (=1050 ksym/s).

Uhh, the decimation values aren’t that restricted.

They must be EVEN and in [8, 256].

http://gnuradio.org/doc/doxygen/classusrp__standard__rx.html#ebd585937cec3e60f1167cddd100f308

Eric

irene159 wrote:

- rational_resampler_base_xxx()

I am not sure of which one I should use. Does somebody know the difference

between them?

I’ve read that rational_resampler_xxx() is a high-level wrapper for

gr.rational_resampler_base_xxx but I don’t understand what it means.

As Eric said, you can use any even decimation rate.

If you need to resample, you’re better off using the

blks2.rational_resampler block which is a convenient wrapper for

gr.rational_resampler.

Tom