Sorry I forgot to edit the title in the last post.

On Thu, Jul 1, 2010 at 10:28 AM, Tuan Ta <[email protected]> wrote:

> Hello all,
>
> I am working in the same team with Shalabh. We have been trying to figure
> out this problem for a week now and so far we found nothing concrete. We
> have been looking in every corner of the archive. We found some people who
> had similar problem with us (some dated back to 2008!) but no one had a
> clear explanation.
>
> We found in some post that if the sampling rate of our signal is too small,
> which leads to small bandwidth, our signal might get lost in the filtering
> stage due to some frequency offset of the filter. Increasing the sampling
> rate (e.g. 500k -> 1M) indeed improved the reception ratio at the receiver.
>
> We managed to successfully send a whole file (small) using gmsk, sampling
> rate 1.5M, payload size 3 bytes. However, if we increased the payload size
> (say 100 bytes), then we still got packet lost even at 1.5M. This leads to
> another question: how does the size of the payload impact the reception at
> the receiver?
>
> Since all packet lost occurred at the end of transmission (always last
> packets that got lost), we suspected some kind of buffer overflow. But
> frankly we don't know how to check that. We printed out the size of the
> message buffer at the receiver USRP after every rx_callback but nothing
> seemed to be out of the ordinary.
>
> This problem has been really bugging us. Since we are new to both the
> hardware and software, we might miss some crucial details. We will be very
> thankful if anyone can provide some pointers / suggestions / explanations
> that help us clarify this issue.
>
> Thank you,
>
> Tuan
>
>
> Message: 13
>> Date: Mon, 28 Jun 2010 19:25:01 -0400
>> From: Shalabh Jain <[email protected]>
>> Subject: [Discuss-gnuradio] benchmark_tx/rx with RFX 2400
>> To: [email protected]
>> Message-ID:
>>        <[email protected]>
>> Content-Type: text/plain; charset="iso-8859-1"
>>
>> Hello,
>>
>> I am new to the USRP/gnu-radio world, trying to understand the basic
>> working. I have been trying the following 2 experiments which I've not
>> been
>> able to successfully debug or find documentation for. Can somebody help me
>> out to explain what is going wrong or point to some useful references.
>>
>> The hardware I am using is the RFX2400 daughterboard with USRP and a
>> 2.4-2.48G antenna. I'm using two separate boards in receive and transmit
>> configuration.
>>
>> 1. I tried the digital/benchmark_tx/rx.py provided with the examples
>> (3.2.2
>> tarball).  I tried to read from a simple ascii file containing a sequence
>> of
>> numbers repeated 1000 times. I've tried different options modulation
>> schemes
>> but the problems persist. Two main problems in this example
>>     a) It seems that using the default gmsk modulation, although 179
>> packets are sent, only 147 are received. Essentially, the last packet is
>> always received in error and the end 15% of the file is always lost.
>>     b) The transmission seems to be very sensitive to the bit rate
>> specified. i.e. there are several combination of big-rates at which the
>> receiver receives nothing.
>>
>> 2. I tried so send a simple tone across by hooking a sine wave generator
>> to
>> the USRP using the grc. Even that seems the just end up in noise
>> reception.
>> I'm attaching a small snippet of the code generated.
>>
>> I appreciate any help.
>>
>> Thanks
>> Shalabh
>>
>>
>> -------------------------------tx-------------------------------------------
>>                self.gr_float_to_complex_0 = gr.float_to_complex(1)
>>                self.gr_sig_source_x_0 = gr.sig_source_f(500000,
>> gr.GR_COS_WAVE, 100000, .5, 0)
>>                self.gr_sig_source_x_1 = gr.sig_source_f(500000,
>> gr.GR_SIN_WAVE, 100000, .5, 0)
>>                self.usrp_simple_sink_x_0 = grc_usrp.simple_sink_c(which=0,
>> side="A")
>>                self.usrp_simple_sink_x_0.set_interp_rate(256)
>>                self.usrp_simple_sink_x_0.set_frequency(2.44e9,
>> verbose=True)
>>                self.usrp_simple_sink_x_0.set_gain(80)
>>                self.usrp_simple_sink_x_0.set_enable(True)
>>
>>                self.connect((self.gr_float_to_complex_0, 0),
>> (self.usrp_simple_sink_x_0, 0))
>>                self.connect((self.gr_sig_source_x_0, 0),
>> (self.gr_float_to_complex_0, 0))
>>                self.connect((self.gr_sig_source_x_1, 0),
>> (self.gr_float_to_complex_0, 1))
>>
>> ------------------------------rx-------------------------------------------
>>                self.usrp_simple_source_x_0 =
>> grc_usrp.simple_source_c(which=0, side="A", rx_ant="TX/RX")
>>                self.usrp_simple_source_x_0.set_decim_rate(256)
>>                self.usrp_simple_source_x_0.set_frequency(2.44e9,
>> verbose=True)
>>                self.usrp_simple_source_x_0.set_gain(80)
>>                self.wxgui_fftsink2_2 = fftsink2.fft_sink_c(
>>                        self.GetWin(),
>>                        baseband_freq=0,
>>                        y_per_div=10,
>>                        y_divs=10,
>>                        ref_level=50,
>>                        sample_rate=500000,
>>                        fft_size=1024,
>>                        fft_rate=30,
>>                        average=False,
>>                        avg_alpha=None,
>>                        title="FFT Plot",
>>                        peak_hold=False,
>>                )
>>                self.Add(self.wxgui_fftsink2_2.win)
>>                self.connect((self.usrp_simple_source_x_0, 0),
>> (self.wxgui_fftsink2_2, 0))
>>
>>
>> -----------------------------------------------------------------------------
>> -------------- next part --------------
>> An HTML attachment was scrubbed...
>> URL:
>> http://lists.gnu.org/archive/html/discuss-gnuradio/attachments/20100628/e47fa55f/attachment.html
>>
>>
_______________________________________________
Discuss-gnuradio mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Reply via email to