discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] About benchmark in USRP E110


From: Philip Balister
Subject: Re: [Discuss-gnuradio] About benchmark in USRP E110
Date: Mon, 30 Jul 2012 12:09:59 -0400
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:14.0) Gecko/20120717 Thunderbird/14.0

On 07/29/2012 06:21 PM, Josh Blum wrote:


On 07/28/2012 04:37 AM, ÀîÔÞ   wrote:
Dear all: Now, I am running the benchmark_tx.py and benchmark_rx.py
on the USRP E110s. However, even if I set the bit rate to 100khz and
use bpsk modulation scheme, overflow always occure. I get some
information that benchmark examples now can not work well in the USRP
E110. So, is there some one run benchmakr examples well in the E110
or is there some suggestion to solve this overflow problem.


Well I suspect the bandedge fll in the bpsk chains takes up a little too
much juice. I might suggest experimenting with the gmsk chains.

Anyway, this is expected. benchmark is an example of implementing a mac
layer w/ gnuradio in python. But its not going to yield continuous
high-rate throughput.

When you code up your application, I suggest making heavy use of the
neon SIMD. :-)

I keep begging people to re-implement the benchmark_{tx,rx} flowgraphs using grac and real gnuradio blocks. Once you start digging into the code, you find it is really difficult to work out how and where to optimize the key functions.

Philip



reply via email to

[Prev in Thread] Current Thread [Next in Thread]