Re: [Discuss-gnuradio] Multiplying the inputs with a scalar/vector in my own block

[Pavan Yedavalli]

Hi Marcus,

After doing more analysis, it seems pretty clear that the signal keeps coming down whenever U is printed out, though I'm not sure how to explain the wrong multiplication aspect - perhaps the USRP is doing a multiply on whatever samples it receives, which is fewer than what it needs to run the process in full?

Anyway, I am reading on various mailing lists that my host CPU should be powerful, which I believe it is (Intel® Core™ i7-4600U CPU @ 2.10GHz × 4), but I'm not sure. I guess what are my options for eliminating these underruns (and hopefully the multiplication error and fluctuation that I'm seeing)? Thanks again for the help.

Also, it appears that the master clock rate on the USRP N210 is 100 MHz, which is even higher than what I had at 30.72 MHz. I tried experimenting with that value (bringing it down to as low as 1 Hz), but after 2 Us are displayed at the beginning, it doesn't seem to transmit again. This is the case for basically everything below 30.72 MHz. At the default 100 MHz, it shows a stream of Ls. I'm guessing that adjusting this rate is probably not the right approach.

Thanks.

On Wed, Jun 8, 2016 at 10:34 AM, Pavan Yedavalli <address@hidden> wrote:

Hi Marcus,

Thanks for getting back to me. To address both things you mentioned:

(1) Yes, that was a typo. One of them should have said in0; I made a mistake while copying over; the code was correct, however, and that behavior still existed.

(2) This is interesting. I am seeing Us on the output basically every time. So, this means the block isn't fast enough to handle 30.72 million samples per second? Should I lower that sampling rate? And to what value? How do we know? And I guess what is the "correct" thing to see on the output? This is very likely my problem. Please keep me posted! Thanks.

--

Pavan

--

Pavan