How to use the SPECTRAN V6 via Matlab
Quote from splash on 24/09/2021, 13:02Hello,
is this feature (http Server for MATLB) supported by the (old) Spectran V5 as well?
I noticed that I can select a http Server for Spectran V5 in RTSA-suite but with my setup it did not work.
Could be that my company PC blocks ports, that's why I ask here.Please help me debuggin. Thanks.
Hello,
is this feature (http Server for MATLB) supported by the (old) Spectran V5 as well?
I noticed that I can select a http Server for Spectran V5 in RTSA-suite but with my setup it did not work.
Could be that my company PC blocks ports, that's why I ask here.
Please help me debuggin. Thanks.
Quote from mm_dev on 24/09/2021, 15:48The HTTP connection will work with a Spectran V5 as well. You can easily test that the HTTP connection works by accessing the URL in a browser (when using the /stream endpoint that will start a download that you can cancel, when using /sample it should display some data).
There might be some minor differences in the data format that could require changes in the Matlab code to be processed, not sure on that.
The HTTP connection will work with a Spectran V5 as well. You can easily test that the HTTP connection works by accessing the URL in a browser (when using the /stream endpoint that will start a download that you can cancel, when using /sample it should display some data).
There might be some minor differences in the data format that could require changes in the Matlab code to be processed, not sure on that.
Quote from jhexa on 21/10/2021, 17:19Hello again
You provided an excellent matlab example how to convert streamed raw16 data to I and Q values. How about when using float16 or float32 as a stream format ? Could you also provide example code on that also ?
Hello again
You provided an excellent matlab example how to convert streamed raw16 data to I and Q values. How about when using float16 or float32 as a stream format ? Could you also provide example code on that also ?
Quote from mm_dev on 26/10/2021, 11:02As far as I can tell, it should be sufficient to replace 'int16' with 'float16' or 'float32' in the cast instruction and remove the '/header.scale' part in the lines below.
Also when using float32 also change 'size = header.samples * 4' to 'size = header.samples * 8'.
But I'm not familiar with Matlab programming, just guessing here what that example program does in detail.
As far as I can tell, it should be sufficient to replace 'int16' with 'float16' or 'float32' in the cast instruction and remove the '/header.scale' part in the lines below.
Also when using float32 also change 'size = header.samples * 4' to 'size = header.samples * 8'.
But I'm not familiar with Matlab programming, just guessing here what that example program does in detail.
Quote from DevSF on 26/10/2021, 13:10Hi jhexa,
mm_dev is absolutely correct.
Also have a look at the typecast function of Matlab, by using the typecast to 'single' which should interpret the data as IEEE 32bit floating point values:
https://de.mathworks.com/help/matlab/ref/typecast.html
16bit floating point precision is not supported by this typecast function. Here is an article at the Mathworks blog about half precision values:
https://blogs.mathworks.com/cleve/2017/05/08/half-precision-16-bit-floating-point-arithmetic/
Hi jhexa,
mm_dev is absolutely correct.
Also have a look at the typecast function of Matlab, by using the typecast to 'single' which should interpret the data as IEEE 32bit floating point values:
https://de.mathworks.com/help/matlab/ref/typecast.html
16bit floating point precision is not supported by this typecast function. Here is an article at the Mathworks blog about half precision values:
https://blogs.mathworks.com/cleve/2017/05/08/half-precision-16-bit-floating-point-arithmetic/
Quote from EnthuMan on 07/02/2022, 14:38Hello,
We are looking to play data generated in Matlab using the V6 (generate I,Q samples in Matlab, send it to the hardware and then play it). Please guide on the process.
Regards,
Hello,
We are looking to play data generated in Matlab using the V6 (generate I,Q samples in Matlab, send it to the hardware and then play it). Please guide on the process.
Regards,
Quote from AdminTC on 07/02/2022, 16:35Use the File Source block which supports the import of different IQ standards (you might need to export your file to one of those from Matlab).
Use the File Source block which supports the import of different IQ standards (you might need to export your file to one of those from Matlab).
Quote from EnthuMan on 08/02/2022, 08:14Hello,
Thank you for your reply. Please guide us where to find the file source block (import file).
Thank you,
Hello,
Thank you for your reply. Please guide us where to find the file source block (import file).
Thank you,
Quote from AdminTC on 08/02/2022, 09:46https://v6-forum.aaronia.de/forum/topic/in-out-all-blocks-at-a-glance/
https://v6-forum.aaronia.de/forum/topic/in-out-all-blocks-at-a-glance/
Quote from EnthuMan on 09/02/2022, 13:54Hello,
Thank you. We found the block.
We are new to this product so may I ask if you could point me to article that step-by-step speaks about implementing the design.
Hello,
Thank you. We found the block.
We are new to this product so may I ask if you could point me to article that step-by-step speaks about implementing the design.