Confusion with uhd_fft.py with usrp2

luislavena · September 10, 2011, 8:01pm

Today I ran uhd_fft.py in two usrp and two usrp2.They were in the same
room
and placed at a distance of 2-3 meters from each other(approx) but there
was
much difference in the fft.

Actually there was a peak at 2.432G . Both usrp detected well but there
was
some confusing display from usrp2 under uhd_fft.py. usrp2s also detected
the
peak but there was a significant rise on the left side in both the
usrp2.I
checked with RFX2400 and XCVR2450 both.

                                       Here is the fft for both

usrp2
under uhd_fft.py

http://old.nabble.com/file/p32438830/usrp2_1.png

                                            Also in the figure below

its detecting two peaks.

http://old.nabble.com/file/p32438830/usrp2_2.png

                                                 Also sometimes it

went
like this

http://old.nabble.com/file/p32438830/usrp2_2_1.png

At the same time the display from usrp under usrp_fft.py was perfect.
I am using RFX2400 as well as XCVR2450. Although there is some problem
with
the second usrp.Despite of elevating the gain and changing the positions
the
peak is not getting sharp and high.

http://old.nabble.com/file/p32438830/usrp_1.png

http://old.nabble.com/file/p32438830/usrp_1_2.png

What may be the reason… Placement of the usrp’s or some hardware
fault ?

Sumit Kr.
Research Assistant
Communication Research center
IIIT Hyderabad
India

View this message in context:
http://old.nabble.com/Confusion-with-uhd_fft.py-with-usrp2-tp32438830p32438830.html
Sent from the GnuRadio mailing list archive at Nabble.com.

sumitstop · September 10, 2011, 10:11pm

Hi All,

I’ve been following closely the OFDM- and related TX/RX issues.

If you TX sumthing, build/buy an (directive) antenna for your target
frequency and point it at the RX usrp.
I suggest a Yagi, helix or a parabola for TX:ing, cost ~$10 at your
frequency. Apply to RX the same.
Using a whip is trouble if you do not know…

An antenna sure can be omni to detect the signal, to demodulate it
perfectly is another thing.

Example: TX is a RHCP (right hand circular polarized) signal.
RX it with a VP (vertical polarized) antenna, it wont demodulate
properly.

A user receiving 137 MHz RHCP satellite data posted his ultimate
antenna
image
http://www.poes-weather.com/~patrik/tmp/RHCP-137.jpg

It aint that complicated, trial and error does it.

Regards,
Patrik

sumitstop · September 12, 2011, 2:18pm

Hey Patrik, will using directional antenna solve my problem.
Patrik T.-2 wrote:

complicated, trial and error does it.Regards,Patrik
_______________________________________________Discuss-gnuradio mailing

[email protected]://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Sumit Kr.
Research Assistant
Communication Research center
IIIT Hyderabad
India

View this message in context:
http://old.nabble.com/Confusion-with-uhd_fft.py-with-usrp2-tp32438830p32447361.html
Sent from the GnuRadio mailing list archive at Nabble.com.

sumitstop · September 12, 2011, 2:56pm

FYI,

There is a great lowcost antenna design software called EZNEC
http://www.eznec.com/

Good luck (oh, we do not need to be lucky since we know…)
Patrik

----- Original Message -----
From: “sumitstop” [email protected]
To: [email protected]
Sent: Monday, September 12, 2011 15:16
Subject: Re: [Discuss-gnuradio] RX and TX stuff