[Discuss-gnuradio] gnuradio history loop

[Mr M.H.M.N.D. Herath]

Dear sir/Madam
 If we set history in the block. Can they process loop in each input sample
signal
Neil Herath

-- 
*CONFIDENTIAL COMMUNICATION*
*E-mails from The Open University of Sri Lanka normally contain 
confidential and privileged material, and are for the sole use of the 
intended recipient. Use or distribution by an unintended recipient is 
prohibited, and may be a violation of law. If you believe that you received 
this e-mail in error, please do not read this e-mail or any attached items. 
Please delete the e-mail and all attachments, including any copies thereof, 
and inform the sender that you have deleted the e-mail, all attachments and 
any copies thereof. Thank you.*
___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] PSK31 (from audio to radio)

[Cinaed Simson]

On 09/09/2016 06:40 PM, pen tester wrote:
> Hi,
>Still playing around and I have a question about the value which
> should go in the Polyphase Resampler on the tx side please. Any advice
> on the AM rx too would be helpful :-)

Looking at transmit.

It appears you start with something in memory so call the initial
pressure x - it will be likely be as fast as it can go.

Next is a resampler which bumps up the sampling rate to x*32.

Skipping the filter - which is a can of worms in itself - the next
resampler bumps up the sampling rate to x*32*48.

So before you get to the multiply, you appear to have some serious
plumbing problem.

Also, I wouldn't be calling math functions to calculate tap values on a
filter. I'd recommend calculation those taps by hand inserting the
values in the filter.

I recommend simplifying it by calculating the everything in front of the
first resampler and stick in a file - create a little back pressure.

And then I would move the FIR filter on the side of the resamplers which
you expect to have the highest sampling rate - which makes one of those
resamplers redundant.

Suppose you can suck in the contents of the file on the filesystem at a
sampling rate of 32k. Then you would need a resampler which decimates at
32k and interpolates at 48k - or interpolates at 3 and decimates at 2 -
to match 32k to 48k.

Now the osomo sink is sampling at 1.2 Hz.

My guess is the 1.2 Hz a mistake - you appear to have meant 1.2 MHz (or
1.2e6) in which case you need to interpolate at 25 and decimate at 1 to
match 48k to 1.2M.

Disclaimer: don't confuse me with someone who knows what they're talking
about - I'm just making this up as I go along. It's simply what I would
do to clean it - it may be wrong.


> 
> Transmit
> 
>   * Simply trying make the working PSK31 audio tx to tx over HackRF
> instead using [1] and updates from Tim's response
>   * Screenshot: http://pasteboard.co/1dLjHxMAU.png
> 
> 
> Receive
> 
>   * Simply trying to make the working PSK31 audio rx work using a RTLSDR
> instead based on [2]
>   * The first line is an attempt to take in the AM signal and just pass
> it on to [2]
>   * For the AM rx, I used [3] as a guide
>   * Screenshot: http://pasteboard.co/1dLZWV9mv.png
> 
> 
> [1] https://github.com/tkuester/gr-psk31/
>  (examples/psk31_tx.grc)
> [2] https://github.com/tkuester/gr-psk31/
>  (examples/psk31_rx.grc)
> [3] https://www.csun.edu/~skatz/katzpage/sdr_project/sdr/grc_tutorial3.pdf
>  
> 
> On 6 September 2016 at 05:49, pen tester  > wrote:
> 
> Tim,
>Yes, your graphs are really useful :-) Thank you for responding
> in such detail! 
> 
>Here is what I am up to:
> 
> Transmit side:
> 
>   * I made your changes to your transmit graph [1]
>   * All the changes were fine but I was confused about the use of
> the Polyphase Resampler as I could not get the values to work
> without underflows when executing i.e. "..."
>   * So instead (this is probably wrong) I tried using a Rational
> Resampler with an Interpolation of 1e6 and a Decimation of 48e3.
>   * Updated version: http://pasteboard.co/hmyGOGTE4.png
> 
>   * Either way, the tx graph looks healthier - just need some advice
> on the resampler perhaps?
> 
> 
> Receive side:
> 
>   * I am using the start of this receive graph [2] as it was using
> radio rx already 
>   * and just modifying the demod portion to be more like your rx graph 
>   * Updated version: http://pasteboard.co/hmzrMBAIi.png
> 
>   * I also need advice from anyone who could point in the right
> direction on the AM rx side please?
> 
> 
> [1] https://github.com/tkuester/gr-psk31/
>  (examples/psk31_tx.grc)
> [2] https://github.com/argilo/sdr-examples
>  (tutorial/6_psk.grc)
> [3] https://github.com/tkuester/gr-psk31/
>  (examples/psk31_rx.grc)
> 
> On 4 September 2016 at 12:33, Tim K  > wrote:
> 
> Hey hey!
> 
> Glad to see someone using this. (And that it's apparently still
> working! Goodness... I need to clean this up...)
> 
> I'm giving advice freehand right now, as I don't have my radios
> with me to test this. There are definitely more elegant ways for
> this solution, this is just the quickest I could think of
> without reverse engineering what I wrote forever ago.
> 
> Since this is AM, I think you'll only need the Real port on your
> "Float to Complex" block. You can attach the "Null Source" to
> t

Re: [Discuss-gnuradio] gnuradio history loop

[Martin Braun]

You can use the history inside the block, you can't feed it back through
a GNU Radio connection.

M


On 09/10/2016 06:01 AM, Mr M.H.M.N.D. Herath wrote:
> Dear sir/Madam
>  If we set history in the block. Can they process loop in each input
> sample signal
> Neil Herath
> 
> *CONFIDENTIAL COMMUNICATION*
> /E-mails from The Open University of Sri Lanka normally contain
> confidential and privileged material, and are for the sole use of the
> intended recipient. Use or distribution by an unintended recipient is
> prohibited, and may be a violation of law. If you believe that you
> received this e-mail in error, please do not read this e-mail or any
> attached items. Please delete the e-mail and all attachments, including
> any copies thereof, and inform the sender that you have deleted the
> e-mail, all attachments and any copies thereof. Thank you./
> 
> 
> ___
> Discuss-gnuradio mailing list
> Discuss-gnuradio@gnu.org
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
> 


___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] Upgrading Gnuradio/GRC

[Martin Braun]

This won't help your immediate issue, but to avoid these kinds of
situations, I recommend not installing hand-compiled GNU Radio (or UHD)
into /usr/local, but rather some dedicated directory.

Cheers,
M

On 09/07/2016 05:58 AM, John B. Wood wrote:
> Hello, everyone.  Short of starting with a fresh Fedora or Ubuntu OS
> build what's the quickest way to remove an older Gnuradio build prior to
> installing a later version?  I have reason to believe trying to install
> a newer build doesn't overwrite all the older stuff and can cause
> startup issues with GRC.  I've manually removed directories for an older
> build but haven't seen an uninstall program to remove all
> Gnuradio/GRC/UHD items.  Other than an uninstall program is there a list
> of directories that should be deleted prior to performing a
> Gnuradio/GRC/UHD upgade? Thanks for your time and comment.  Sincerely,
> 
> 
> John Wood
> U.S. Naval Research
> Laboratory
> 
> ___
> Discuss-gnuradio mailing list
> Discuss-gnuradio@gnu.org
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio


___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] PSK31 (from audio to radio)

[pen tester]

Thanks Cinaed for your response.

For everything up to the multiply on the TX side - this is actually from a
working example that TXs using an audio card [1]. So just for a proof of
concept I would like to leave that as-is and just append on TX over Osmocom
Sink. Happy to look at simplifying later, just wanted to get POC for PSK31
over radio for now on the transmit and receive.

Also thanks for spotting my Freq mistake too.

My plan now is to use the fldigi app for testing the rx side while I try
and make the rx side work - just figuring out how to write to a wav file
sink that is accepted by fldigi too... and trying to find some TX examples.

[1] https://github.com/tkuester/gr-psk31/ (examples/psk31_tx.grc)

On 10 September 2016 at 11:40, pen tester  wrote:

> Hi,
>Still playing around and I have a question about the value which should
> go in the Polyphase Resampler on the tx side please. Any advice on the AM
> rx too would be helpful :-)
>
> Transmit
>
>- Simply trying make the working PSK31 audio tx to tx over HackRF
>instead using [1] and updates from Tim's response
>- Screenshot: http://pasteboard.co/1dLjHxMAU.png
>
>
> Receive
>
>- Simply trying to make the working PSK31 audio rx work using a RTLSDR
>instead based on [2]
>- The first line is an attempt to take in the AM signal and just pass
>it on to [2]
>- For the AM rx, I used [3] as a guide
>- Screenshot: http://pasteboard.co/1dLZWV9mv.png
>
>
> [1] https://github.com/tkuester/gr-psk31/ (examples/psk31_tx.grc)
> [2] https://github.com/tkuester/gr-psk31/ (examples/psk31_rx.grc)
> [3] https://www.csun.edu/~skatz/katzpage/sdr_project/sdr/grc_tutorial3.pdf
>
>
> On 6 September 2016 at 05:49, pen tester  wrote:
>
>> Tim,
>>Yes, your graphs are really useful :-) Thank you for responding in
>> such detail!
>>
>>Here is what I am up to:
>>
>> Transmit side:
>>
>>- I made your changes to your transmit graph [1]
>>- All the changes were fine but I was confused about the use of the 
>> Polyphase
>>Resampler as I could not get the values to work without underflows when
>>executing i.e. "..."
>>- So instead (this is probably wrong) I tried using a Rational
>>Resampler with an Interpolation of 1e6 and a Decimation of 48e3.
>>- Updated version: http://pasteboard.co/hmyGOGTE4.png
>>- Either way, the tx graph looks healthier - just need some advice on
>>the resampler perhaps?
>>
>>
>> Receive side:
>>
>>- I am using the start of this receive graph [2] as it was using
>>radio rx already
>>- and just modifying the demod portion to be more like your rx graph
>>- Updated version: http://pasteboard.co/hmzrMBAIi.png
>>- I also need advice from anyone who could point in the right
>>direction on the AM rx side please?
>>
>>
>> [1] https://github.com/tkuester/gr-psk31/ (examples/psk31_tx.grc)
>> [2] https://github.com/argilo/sdr-examples (tutorial/6_psk.grc)
>> [3] https://github.com/tkuester/gr-psk31/ (examples/psk31_rx.grc)
>>
>> On 4 September 2016 at 12:33, Tim K  wrote:
>>
>>> Hey hey!
>>>
>>> Glad to see someone using this. (And that it's apparently still working!
>>> Goodness... I need to clean this up...)
>>>
>>> I'm giving advice freehand right now, as I don't have my radios with me
>>> to test this. There are definitely more elegant ways for this solution,
>>> this is just the quickest I could think of without reverse engineering what
>>> I wrote forever ago.
>>>
>>> Since this is AM, I think you'll only need the Real port on your "Float
>>> to Complex" block. You can attach the "Null Source" to the Imaginary port.
>>> Do a quick check to make sure the real signal's not exceeding +/- 1... you
>>> might want to attenuate it to +/- 0.7 for now, someone else should comment.
>>>
>>> As far as the transmit side goes, I'd leave everything up to the
>>> "Multiply" block untouched. Consider that your "audio input" to the radio.
>>> After that...
>>>  - Multiply Const (set to 0.7 for safety)
>>>  - Polyphase Resampler (float/float) set to (1e6 / 48e3) - This gets
>>> your sample rate up to radio speeds
>>>  - Float to Complex (in the Real port, Null Source to the Imag port) -
>>> This makes your signal "complex AM"
>>>  - Send the Complex stream into your Osmocom Sink (set to 1e6 sps) - And
>>> transmit!
>>>
>>> If I remember correctly, some radios don't like working at non-integer
>>> sample rates. (The hackRF may be ok though...)
>>>
>>> Play gently with your TX gain until you can see it on another radio.
>>> (IIRC the HackRF is simplex... perhaps an RTL?) Use AM to demodulate.
>>>
>>> I've been meaning to do a write up on this for a while. If I ever get
>>> around to it, I'll make sure to CC you.
>>>
>>> - Tim
>>>
>>>
>>> On Fri, Sep 2, 2016 at 11:55 PM, pen tester 
>>> wrote:
>>>
 Hi,
 I am trying to get a simple working example of PSk31 working
 (transmitting with a hackrf and receiving with a rtlsdr dongle). Here is
 what I have so far.
>>