Hi Marcus/Muller,
I was trying a simple flow graph to implement what you said earlier :
“
If it were *my* problem, I'd probably have a couple of vector sources of alternating 1s and 0s, interpolate them up to the desired sample rate
X the switching interval, and multiply the stream going into the USRP by that stream of 1s and 0s.
“
To make things simpler, I am trying:
Vector source (0,1) multiplying with a sine wave. As an out put I want 1 period of sine wave (say 1kHz frequency) and the next period to be zero. I believe I need to set the sample rate and match this period with
the vector source.
Can you just help be in deducing this please. Having hard time to sort this out
☹
Thanks
Ayaz
The Selector block comes from the "deprecated" category, and that for good reason: Don't use it, it causes a huge teardown/setup effort behind the scenes, and not all is guaranteed to work.
Follow Marcus' advice with the 0/1 multiplication and offset tuning.
On Tue, 2018-07-17 at 23:06 +0000, Ayaz Mahmud wrote:
Hey Marcus,
I think the Selector block works well with my setup.
But I need a solution with which I can do a dynamic change in the input/output index. Can you let me know how is this possible
Ayaz
Hi Marcus,
The situation still is same.
By browsing the archive I found some switching of blocks with SELECTOR. Is there any example that will be helpful on how this block actually functions ?
Thanks
Ayaz
On 07/16/2018 02:05 PM, Ayaz Mahmud wrote:
Hi Marcus,
The issue is even I put the vector source toggling with 1 and 0s it transmit a signal. So I need something that will not transmit at all and switch between these 2 radio chains.
Thanks,
Ayaz
You're probably seeing the residual LO leakage during periods of "no transmit".
You can try using offset-tuning to move the leakage out-of-band:
Use:
uhd.tune_request(frequency,offset)
In the UHD sink for the "frequency" field.
On 07/14/2018 06:40 PM, Ayaz Mahmud wrote:
Hi Muller / Marcus,
I got the testing right now, but unfortunately it doesn’t work while I put it in my original file.
Here is the attached .grc where I have used vector source as input and multiply a period with 1 and then with zeros on one radio channel and opposite on the other.
Can you please have a look, and check if I am doing anything wrong ?
Ayaz
You're asking us to debug a very complex signal flow.
Why don't you try something a *lot* simpler to get the TX-flow-switching-logic correct first, and then work up from there.