# Polyphase clock synchronization issue

**URL:** <https://www.ruby-forum.com/t/polyphase-clock-synchronization-issue/241114>\
**Category:** GNU Radio\
**Created:** [May 13, 2014, 5:14pm UTC](https://www.ruby-forum.com/t/polyphase-clock-synchronization-issue/241114 "2014-05-13T17:14:03Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Detlef\_R](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/58f4c7/32.png) [@Detlef\_R](https://www.ruby-forum.com/u/Detlef_R)\
**Post date:** [May 13, 2014, 5:14pm UTC](https://www.ruby-forum.com/t/polyphase-clock-synchronization-issue/241114/1 "2014-05-13T17:14:03Z")

</div>

Hi all,

I have a problem here with polyphase clock synchronization block when i  
set  
the output samples per symbol to one the constellation have much more  
ISI  
and noise and when output sps value is 2 the constellation have no noise  
and no ISI.I am using the firdes.root\_raised\_cosine filter design in  
polyphase clock synchronization block with the follwoing parameters

Polyphase clock synchronization block parameters  
filter size =32  
firdes.root\_raised\_cosine(nfilts,1.0,1.0/(nfilts_sps), rolloff,  
int(11_sps\*nfilts)) with sps=2 nfilts=32 rolloff=0.7

My flow graph picture is attached with this mail.  
please tell me if there is any problem in my parameters (I m using QPSK  
using the chunks to symbol block with const.points() function for mod  
and  
constellation decoder block with const.base() function in the  
constellation  
object parameter for demod)  
If there are any errors in my flowgraph parameters please guaid me.

Regards,  
Irfan U.

---

<div class="post-metadata">

**Author:** ![Irfan\_U](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/i/57b2e6/32.png) [@Irfan\_U](https://www.ruby-forum.com/u/Irfan_U)\
**Post date:** [May 14, 2014, 4:34pm UTC](https://www.ruby-forum.com/t/polyphase-clock-synchronization-issue/241114/2 "2014-05-14T16:34:19Z")

</div>

On Tue, May 13, 2014 at 11:11 AM, Irfan U.  
[removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)wrote:

> firdes.root\_raised\_cosine(nfilts,1.0,1.0/(nfilts\*sps), rolloff,  
> Irfan U.

You have the output SPS set to 2. Set that to 1. Basically, at 1 sps  
out,  
you’ll get the sampled symbol that you then pass to the Costas loop for  
phase and fine frequency recovery. If you output at 2 sps like you have  
now, then there’s nothing that takes that down to the single bit or  
symbol  
sample that you want. The 2 sps output is specifically designed to go  
into  
an equalizer as the next stage, which will take 2 sps in and produce the  
1  
symbol out after equalization.

Tom