Problem with USRP interpolation

GNU Radio
1

Hi,

Recently, I’m trying to do CDMA commnication at 1.28Mchip/s with USRP
and
I’ve met a problem.
The sample rate of DAC on USRP is 128M, so samples avialabe for each
chip is
128M / 1.28M = 100.
The interpolation rate of USRP must be [4, 512] and multiple of 4. So
the
only choice for me seems to be setting the software interpolation rate
to 5
and the interpolation rate of USRP to 20 (multiple of 4). 5 * 20 = 100.
But
the software interp is still too high for the root_raise_cosin interp
filter
(interp=5, len(taps)=16), and a lot of “Uu” comes out. However, the
software
interp cannot be reduced further.

I wonder if the sample rate of DAC can be change to mutiple of 2, not 4.
In
that case, I can set the software interp rate to 2 and interp rate of
USRP
to 50, for which the burden of interp filter can be reduced.

Right now, I have to use 1.3333Mchip/s. The software interp rate of 2
works
well. But 1.28Mchip/s is exactly what I want.

Thank you :slight_smile:

2

hanwen wrote:

root_raise_cosin interp filter (interp=5, len(taps)=16), and a lot of
“Uu” comes out. However, the software interp cannot be reduced further.

I wonder if the sample rate of DAC can be change to mutiple of 2, not
4. In that case, I can set the software interp rate to 2 and interp
rate of USRP to 50, for which the burden of interp filter can be reduced.

With the current FPGA build, the interpolation must be a multiple of 4.
If you wanted to go to the effort, you could build your own version
which would:

  • have the digital upconverter in the FPGA instead of using the one in
    the AD9862
  • run the data to the AD9862 at 128 MHz instead of 64

Right now, I have to use 1.3333Mchip/s. The software interp rate of 2
works well. But 1.28Mchip/s is exactly what I want.

You might be better off using an external clock into the USRP which is a
more convenient multiple.

Matt

3

hanwen wrote:

root_raise_cosin interp filter (interp=5, len(taps)=16), and a lot of
“Uu” comes out. However, the software interp cannot be reduced further.

The other option is to figure out why this isn’t running fast enough.
Is the USB bus too slow or is it your processor? What sort of processor
are you using? You shouldn’t have any problem at this data rate.

Try this:

usrp_siggen.py -d 20

If that underruns, your USB bus is not fast enough. Otherwise it is
your processor.

Matt