[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Discuss-gnuradio] Having a higher samples per symbol (greater than
From: |
Jason Uher |
Subject: |
Re: [Discuss-gnuradio] Having a higher samples per symbol (greater than 7) with lower bitrate |
Date: |
Sun, 12 Jul 2009 11:17:31 -0500 |
=> the maximum samples per symbol value that benchmark_tx has is 7. What
> happens if I change it to a higher value and reduce the bitrate such that
> the usrp constraints are still met. I read somewhere that having a sample
> per symbol value greater than 7 is impractical. Why?
>From a strictly 'seat of the pants' point of view, it seems to work
fine. You can edit the array (I think in pick_bitrate.py) that limits
the search space for bitrates to a sps less than 7. You can add your
own like, 50 and 60 for example, and pick_bitrate will then allow for
you to use the very low data rates associated.
As for why this might be bad, I have no idea, we did it mainly as a
'look what we can do' for testing out a large range of bitrates. When
we took experimental data over a range of SNRs using a channel
emulator to set the SNR via additive noise we found that the empirical
measurements very closely matched the theoretical limits expected from
BPSK.
Jason