RE: The better mapper and the frequency of the "S" header bit

["Devi, Sriram" <sriram.devi@xxxxxxxxx>]

 

Hi Gareth,

   LTE-TDD radio frame (10ms) consists of 10 sub-frames (1ms each) with specific sub-frame types. In this case, if sub-frame boundary is used as

the sync point, the current sub-frame type (U/D/S) information is not known. Hence, radio frame boundary can be used as  the sync point.

 

Thanks,

Sriram

 

 

From: stds-1904-3-tf@xxxxxxxx [mailto:stds-1904-3-tf@xxxxxxxx] On Behalf Of Gareth Edwards
Sent: Friday, August 21, 2015 6:26 AM
To: stds-1904-3-tf@xxxxxxxx
Subject: The better mapper and the frequency of the "S" header bit

 

Hello all,

 

In the meeting this morning there was some discussion over p7 of tf3_1508_korhonen_cpri_better_mapper_3a.pdf

 

http://www.ieee1904.org/3/meeting_archive/2015/08/tf3_1508_korhonen_cpri_better_mapper_3a.pdf

 

with Richard Maiden suggesting that the hyperframe boundary was too often and the radio frame boundary was the key synchronization point

 

I had a think about this after the call and agree with Richard on the hyperframe boundary being suboptimal for the ‘S’ bit occurrence – my only uncertainty is whether the 10ms frame or the 1ms subframe boundary is the more optimal sync point for the ‘S’ bit to appear in – probably the 10ms frame but it’s worth discussing the alternative.

 

I would guess that the reason that CPRI originally specified hyperframe boundaries was to provide a more frequent source of commas for the serdes synchronization in the physical layer, but with Ethernet providing our transport we don’t care.

 

Thanks

Gareth