Could this output be a problem with my Hardware or the python codes?

GNU Radio
1

Hey all,

May be my question could be out of topic as it is can be a hardware
problem, so excuses.

The Thing is…I am using USRP with XCVR2450 Tranciever to develop a
doppler radar(in duplex mode).
So, I was checking out my devices.

When I use USRP_fft.py just to see the environment signals without
enabling
my USRP_siggen.py, I see a certain signal level.
Then when i start to transmit and recieve in duplex mode running these
two python codes simultaneously in different shell scripts; the
signal(FFT)
level
I am recieving decreases too much. Normally, I should get a more strong
recieved signal rather since I am transmitting.

So, does anyone has experience in this stuff ? If so, what do you think
is
the problem? could it be a problem with my Trancievers.

Thank you!

Bruhtesfa

2

On Nov 3, 2008, at 8:55 AM, Bruhtesfa E. wrote:

I am using USRP with XCVR2450 Tranciever to develop a doppler
radar(in duplex mode).

When I use USRP_fft.py just to see the environment signals without
enabling my USRP_siggen.py, I see a certain signal level.
Then when i start to transmit and recieve in duplex mode running
these two python codes simultaneously in different shell scripts;
the signal(FFT) level
I am recieving decreases too much. Normally, I should get a more
strong recieved signal rather since I am transmitting.

Bruhtesfa - Are you transmitting and receiving in the same frequency
band using the same daughtercard (even same antenna?)? If so, what
you’re seeing is normal IME … something like 20 dB of attenuation on
RX when in transceiver mode when compared with just receiver mode …
I don’t know if this behavior is “correct” but it’s what I’ve
experienced. If this is not your configuration, could you elaborate
on the script arguments? - MLD

3

On Mon, Nov 03, 2008 at 02:55:53PM +0100, Bruhtesfa E. wrote:

my USRP_siggen.py, I see a certain signal level.
Bruhtesfa
You need to run the Tx and Rx in the same graph. The XCVR2450 shared
resources (e.g., LO) between the Tx and Rx side, and if you run the Tx
and Rx in different processes they don’t know about each other and
clobber each other.

Eric

4

Are you trying to simultaneously transmit and receive with the
XCVR2450? Because it doesn’t do that – it does not do FDD. The
RFX-series will work in full duplex mode, but you still need to put them
on different frequencies.

If you really need to do this with the XCVR2450, then you will need two
of them.

Matt

5

Hey,

I have two XCVR2450 Trancievers.I have also two wire antennas connected
to
each.
So,I have separate transmit and recieve paths, but I am using a single
USRP.

I am using the transmiter and reciever daughterboards at almost the same
frequency.
I am transmitting at 2.45GHz and I expect to recieve a reflected signal
at a frequency of 2.45GHz + Doppler shift, so that I will detect the
doppler
shift.

So, what do you think is the problem?

Thanks 4 ur help !

Bruhtesfa

6

Hi Matt/Micheal

I have two XCVR2450 Trancievers.I have also two wire antennas connected
to
each.
So,I have separate transmit and recieve paths, but I am using a single
USRP.

I am using them the transmiter and reciever boards at almost the same
frequency.
I am transmitting at 2.45GHz and I expect to recieve a reflected signal
at a frequency of 2.45GHz + Doppler shift.

So, what do you suggest?

Thanks!

Bruhtesfa

7

I have one USRP and 2 pairs of BASIC TX and BASIC RX daughter cards. Can
i
use the transmitter and the receiver at the same time?
Is there any python program implementing this?

Ali

8

Hey Paul,

Ya! you are correct, the isolation is a problem.But what is curious is
when
i operate the transmitter and reciever simultaneously,the signal level
of
the reciver decreases too much(cancels out) . If some of the signal from
the
transmitter antenna is also recived by the reciver antenna, I think the
signal recived should increase in amplitude, because the frequency of
the
signals sent and recived are almost the same(except the small doppler
shift). But, what occurs is the opposite.

Bruhtesfa

9

Are you aware that the isolation between the transmit and receive
channels
of USRP daughterboards is generally < 40 dB? (This fact made it
impossible
for me to use the Transmit channels in my application.) It is quite
challenging to have transmit and receive channels operating
simultaneously,
sharing a power supply, and without much shielding. I suppose we should
be
surprised that the isolation is as good as it is. As others have
commented,
the expectation is that your transmit and receive channels will be well
separated, otherwise, crosstalk can overwhelm the receiver channels.
Paul M.

10

On Fri, Nov 07, 2008 at 11:19:40AM +0100, Bruhtesfa E. wrote:

Bruhtesfa
You may be overdriving the input resulting in some kind of non-linear
behavior. Have you plotted the received signal? Does it look like
what you think you’re sending?

You say that you are trying to receive a doppler shifted echo. Given
your physical setup (antennas, their pattern and their spacing, the
distance to the object that you expect the signal to reflect from, its
velocity and radar cross section), of the power received at the Rx
antenna – according to theory – how much of the power will come
directly from the Tx antenna, and how much will be reflected off of
the moving object? Assume a free space model for simplicity.

Eric

11

Hey Eric,

It is like this.
(1). when I am not transmitting anything rather just using the USRP as
receiver, I see the environment signal level using USRP_fft.py in
oscilloscope mode.
(2). When I start to transmit and recieve simultaneously, then the
signal
recived is the environment signals + the transmitted signal reflected
from
the target + the signal leaking from the transmitter to reciver antenna.

So, I expect that the signal I receive in (2) should be much higher in
amplitude than the one I receive in (1). But, what I am getting while
doing
this experment is the opposite.
Actually,I preffered to transmit and receive simultaneously using the
USRP
because my target is within a distance of 10meter(I am using the doppler
radar for human movement detection).
But now because of this isolation problem, I am thinking to Implement
my
radar as pulse radar such that I will transmit for a very short time
then
stop transmiting and start recieving, then transmit etc…
So, what do you suggest me?

Thanks!
Bruhtesfa