discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] Using GNU Radio to accumulate FFT samples over a


From: Terrance Pat
Subject: Re: [Discuss-gnuradio] Using GNU Radio to accumulate FFT samples over a constant time duration before writing to disk
Date: Fri, 8 Sep 2017 12:16:26 -0700

Oops, I sent the incorrect flight_recorder.grc flow graph without the
integration block. Here's the one I referred in the email.

Thanks,
--
Terrance

On Fri, Sep 8, 2017 at 12:04 PM, Terrance Pat <address@hidden> wrote:
> Hi everyone,
>
> I am working on a project that requires the use of the USRP B210
> and GNU Radio to act as a receiver and record signals being
> transmitted from a RF source.
>
> Initially, I thought it would be as simple as just taking the UHD_USRP
> source and piping it to the file sink. However, it becomes apparent
> the file holding the raw data quickly becomes very large; on the order
> of gigabytes in a matter of minutes. It is expected that the receiver
> will be collecting data for several hours and our disk space in the
> receiver system is rather limited.
>
> One person suggested having the software
> accumulate or integrate the data samples after they've been FFTed for
> a predetermined duration of integration time before writing to disk to
> reduce the amount of overhead.
>
> In the attached flight_recorder.grc file, I insert an integration
> block between the USRP source and the file sink to do this, but using
> the flight_recorder_playback.grc to play back the signal shows that
> the sample rate is significantly reduced. Initially I thought
> decimation had something to do with the time duration the integration
> is done with the samples, but that was not the case and further
> research showed decimation is defined as reducing the sampling rate.
>
> I also tried the same approach in the attached flight_record_fft.grc
> flow graph with the difference being I pipe the raw data to the FFT
> block before running it through the integration block and writing to
> disk.
>
> Accumulating the data for some integration time should increase my
> signal to noise ratio, but that is not happening with my current
> approaches and there doesn't seem to be a way to tell the integration
> block to integrate over a constant time duration. Perhaps I
> misunderstood how the integration block works in GNU Radio.
>
> Is there another approach using standard blocks in GNU Radio for
> accumulating the FFTed data samples from a USRP source over a constant
> integration time before writing to disk?
>
> I am a novice when it comes to GNU Radio and any assistance is much 
> appreciated.
>
> Thanks,
> --
> Terrance

Attachment: flight_recorder.grc
Description: Binary data


reply via email to

[Prev in Thread] Current Thread [Next in Thread]