discuss-gnuradio
[Top][All Lists]
Advanced

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

[Discuss-gnuradio] [ Implementing a Phase Shift block ]


From: Ashish Raste
Subject: [Discuss-gnuradio] [ Implementing a Phase Shift block ]
Date: Thu, 17 Jan 2013 15:05:52 +0800

Hi Marcus,


> I'm working on synchronising two USRPs (connected via a MIMO) and have
> calibrated the time-offset which occurs between them for a given set
> of signal generator's frequency, USRPs frequency and the sampling rate.
>
> My next step is to do a phase-compensation for one of the USRPs so
> that both are aligned in the time-domain. I read from this
> <http://www.ettus.com/content/files/kb/mimo_and_sync_with_usrp.pdf>
> document that for Phase-shifting, we need to use MRC algorithm or
> Periodic Calibration. Also I saw that a "Phase Shift" block has been
> used in that document which I guess was developed by the author who
> wrote it, as I could not find it in the GRC.
>
> I'm willing to develop one such simple phase-shift block. Can anyone
> suggest me how to approach in implementing it i.e in terms of any
> algorithm's reference that you can suggest to achieve this purpose?
>
> Note: I want my Phase-Shift block to work in real time i.e it should
> be able to adjust the phase-offset of one of the signals so that that
> signal gets aligned with the other in the time-domain.

The attached demonstrates phase-shifting using a simple complex multiply
block.

If you want to do coarser phase shifts (large multiples of 180deg), you
can use a delay block.


Presently I'm calculating the delay between two synchronised signals coming out of the USRP source (connected to a signal generator) by recording their samples and finding the offset between their first peak or zero-crossing.

Now if we want to time-align them in real-time, how can we implement a delay block that calculates the delay each time the grc flow is run and set the delay_value in the block accordingly so that they are time-aligned? i.e I don't want to record the samples rather I'd like to know how to find the delay and shift any of the two signals so that they get aligned?

Best,
 

reply via email to

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