Link Improvement via Feedback

Hello all,

We’re interested in using USRP2s and GNURadio for a wireless link, and
adding a cognitive engine to it to improve link performance (PHY only).

Currently, we are using benchmark_rx.py and benchmark_tx.py (from
digital
folder). Each USRP is connected to a seperate laptop via Ethernet.

We can change knobs (eg: transmit power) at the transmit side (Node1)
now
during transmission. We are working on getting meter values (metrics
such as
SNR, BER, RSSI, etc.) from the receiver (Node2) and then passing them
back
to Node1.

We’ve noticed that there is digital-ber rx scripts, can this be used
with
the digital/benchmark_tx.py in order to transfer a file? We did not see
support for a specified file in the digital-bert/benchmark_tx.py script.

What kind of set-up have people used for Cognitive radio feedback? If
anyone
has done this before and has suggestions, please let us know! Thanks in
advance.

On Thu, Sep 30, 2010 at 3:49 PM, Daniel A. [email protected]
wrote:

What kind of set-up have people used for Cognitive radio feedback? If anyone
has done this before and has suggestions, please let us know! Thanks in
advance.


Daniel A.
Graduate Student
Electrical and Computer Engineering
Virginia Tech

Daniel,
Since you’re at VT, I’m assuming you’re familiar with my dissertation
and/or book (warning, plug coming) Artificial Intelligence in Wireless
Communications? It deals with using GNU Radio as a platform for
cognitive radio. I didn’t deal with feedback; I was interested in
having each radio sense its own meters and make decisions off those.
So it doesn’t address your specific issues, but might be helpful,
anyway.

Tom