Re: Doubts About the Polyphase Clock Sync Block

[Jeff Long]

1. Given that set_damping_factor() requires the value to be in [0.0, 1.0], 2 * d_nfilters looks quite wrong. I wonder if it was supposed to be 2.0 / d_nfilters. This is probably overridden by a call to set_damping_factor() in a typical flowgraph.

I'm not familiar enough with the block to comment on 2 and 3 without some study. Please submit an issue at https://github.com/gnuradio/gnuradio/issues so we don't lose this info.

On Sat, Apr 29, 2023 at 3:56 AM Saurav Roy <sauravroy@iisc.ac.in> wrote:

Dear all,

I have a few questions about the Polyphase Clock Sync Block in reference to https://github.com/gnuradio/gnuradio/blob/master/gr-digital/lib/pfb_clock_sync_ccf_impl.cc

1. line no 67:    d_damping = 2 * d_nfilters;   -->  d_nfilters=32, should this not be around 0.707
   
   
   
2. line no 366:  based on the 'time_est' tag, determine the filter index, d_k. But, this completely bypasses the TED and the control loop as in line no 374 d_filtnum = (int)floor(d_k);
   then why do we need this Polyphase Clock Sync Block at all?
   
   

3. How does the main loop take care of which input samples are selected for the determination of d_error and d_k such that all the input samples come from the same input symbol?

With thanks and regards 

Saurav