Communication problems between 2 USRP's

Hi all,
I want to transmit the data between two USRP’s and make them communicate
with each other. But I guess the packets are not being received
properly…
My USRP daughterboard model is XCVR2450. When I am running
./benchmark_tx.py -f 2.435G, it is starting to transmit the packets but
a small error occured given below,

linux; GNU C++ version 4.8.4; Boost_105500;
UHD_003.009.git-144-g407e3086

Using Volk machine: avx_64_mmx
– Opening a USRP1 device…
– Using FPGA clock rate of 64.000000MHz…

No gain specified.
Setting gain to 7.500000 (from [-20.000000, 35.000000])

UHD Warning:
The hardware does not support the requested TX sample rate:
Target sample rate: 0.050000 MSps
Actual sample rate: 0.250000 MSps

Symbol Rate: 25000.000000
Requested sps: 2.000000
Given sample rate: 250000.000000
Actual sps for rate: 10.000000

Requested sample rate: 50000.000000
Actual sample rate: 250000.000000
…U…terminate
called after throwing an instance of ‘uhd::runtime_error’
what(): RuntimeError: usb tx2 transfer status: 1
Aborted (core dumped).

But towards the receiver side, I gave the command ./benchmark_rx.py -f
2.435GG,it has shown the result given below with n_right=0.

linux; GNU C++ version 4.8.4; Boost_105500;
UHD_003.009.git-144-g407e3086

Using Volk machine: avx_64_mmx
– Opening a USRP1 device…
– Using FPGA clock rate of 64.000000MHz…

No gain specified.
Setting gain to 56.250000 (from [0.000000, 112.500000])

UHD Warning:
The hardware does not support the requested RX sample rate:
Target sample rate: 0.050000 MSps
Actual sample rate: 0.250000 MSps

Symbol Rate: 25000.000000
Requested sps: 2.000000
Given sample rate: 250000.000000
Actual sps for rate: 10.000000

Requested sample rate: 50000.000000
Actual sample rate: 250000.000000
ok = False pktno = 53034 n_rcvd = 1 n_right = 0
ok = False pktno = 24 n_rcvd = 2 n_right = 0
ok = False pktno = 35 n_rcvd = 3 n_right = 0
ok = False pktno = 44 n_rcvd = 4 n_right = 0
ok = False pktno = 46 n_rcvd = 5 n_right = 0
ok = False pktno = 46 n_rcvd = 6 n_right = 0
ok = False pktno = 3872 n_rcvd = 7 n_right = 0
ok = False pktno = 12304 n_rcvd = 8 n_right = 0
ok = False pktno = 49 n_rcvd = 9 n_right = 0
ok = False pktno = 50 n_rcvd = 10 n_right = 0
ok = False pktno = 54 n_rcvd = 11 n_right = 0
ok = False pktno = 200 n_rcvd = 12 n_right = 0
ok = False pktno = 63 n_rcvd = 13 n_right = 0

My USRP can access in the range of 2.4GHZ to 5GHZ. Please advice as to
what I can do. Thanks

Regards,
John

On Thu, Jul 30, 2015 at 6:30 PM, John G. [email protected]
wrote:

Actual sample rate: 0.250000 MSps

called after throwing an instance of ‘uhd::runtime_error’

Actual sample rate: 0.250000 MSps

ok = False pktno = 35 n_rcvd = 3 n_right = 0

My USRP can access in the range of 2.4GHZ to 5GHZ. Please advice as to
what I can do. Thanks

Regards,
John

John,

There’s lots of things that can go wrong here. You didn’t set the gains
or
transmit amplitude. Do you know if you have a significant frequency
offset
between your two radios? You might need to adjust that.

The fact that you are getting “ok = False” means that the receiver saw a
packet, but there were too many errors and the CRC check failed. Note
there
is no FEC applied to the benchmark scripts. So this could be due to low
SNR, a bad channel, or, frankly, many other possibilities.

Tom

Hi Tom,
So can you please advice what I can do next for them to be right?

Regards,
John

On Sat, Aug 1, 2015 at 1:52 PM, John G. [email protected]
wrote:

Hi Tom,
So can you please advice what I can do next for them to be right?

Regards,
John

Too many things; you’ll have to go through it step-by-step. Best thing
to
start with is to transmit and use uhd_fft at the receiver end to look at
the signal. Make sure everything looks ok and try to find a gain setting
that maximizes the SNR.

Tom

Hi Tom,
I have actually added a jpg file which shows the output when given the
command of uhd_fft. Actually,my experiment is to transmit different data
packets to the receiver with minimized errors with ./benchmark_tx.py -f
2.435G and ./benchmark_rx.py -f 2.435G as commands . May I please know
of how can to set the gain in this picture and also how this picture is
useful for the reception of data packets. Please help as I am new to
using GNU Radio and USRP. Thank you.

Regards,
John

Hi Tom,
Attached is the picture that I have sent you previously. I have somewhat
adjusted the receiver gain for the signal to be a flat one. As you have
mentioned before about how to apply the gain settings to the benchmark
scripts, how do you suggest I do that? Also, how do I change the
transmitting gain setting. If I do this, will I be able to receive the
data packet correctly?. I mean, I need to get n_right=1, that is the CRC
check needs to be right. Please suggest as to what I can do.
Thanks.

Regards,
John

On Mon, Aug 3, 2015 at 5:09 PM, John G. [email protected]
wrote:

John

Attachments:

Hi John,

That PNG you sent doesn’t show any signal that I can tell, which
indicates
that your gain settings are either too low to not see it or too high
causing nonlinear effects. In this case, I’d guess the former. You can
adjust the gain directly on the uhd_fft app there with the RX Gain
slider.
Play with that to see if you can see the signal better. Then use the
gain
setting as an option to the benchmark scripts. You may also want to
adjust
the transmit gain setting, as well.

But the issue here isn’t GNU Radio or USRP. This is now radio. You have
to
have good enough signal power to receive data, and so you need to adjust
the gains of the radio to make that happen.

Tom

On Fri, Aug 7, 2015 at 5:53 PM, John G. [email protected]
wrote:

Regards,
John

Attachments:

The problem is that I still don’t see a signal in that screen shot
you’re
sending me.

To adjust the gains and amplitudes, in the benchmark scripts, use the
‘-h’
flag to get help to see what options you can pass to it.

I cannot tell you if you will get the packets received correctly, and
certainly not all of them will be. There is no error correction on that
link, so even 1 bit wrong in a packet will force that entire packet to
be
marked as false because the CRC check will fail.

In the end, these aren’t so much GNU Radio questions but radio
questions,
so then it’s how you work with GNU Radio to address them. You have to
work
on sample rate, frequency offset, gains, etc. etc. The tools are all
there
with the sliders in the GUI apps and the command-line arguments for the
benchmark programs.

Tom

Hi Tom,
Thanks for the reply. I have been trying to import ./benchmark_tx.py -f
2.435G and ./benchmark_rxpy -f 2.435G commands and I have not yet been
able to resolve the problem. I am attaching the files of the benchmark
scripts. What do I need to change in order for my packet to receive
correctly? Please help as I am new to GNU Radio.

Regards,
John