|
Posted by on February 14, 2007, 2:11 pm
If you were Registered and logged in, you could reply and use other advanced thread options
Hi,
I was wondering if there is a difference between the 8b/10b paper/
patent by Widmer and Franaszek and the one adopted by FC-PH and 802.3
specs. This is mainly with regards to the decoder and the disparity
calculations.
The six bit sub-block 000111 and four bit sub-block 0011 are
considered disparity neutral in the paper but are considered to be
positive disparity by the standards. This poses no conflict when doing
an encoder as the coding does result correctly. However, in the case
of a decoder, it does pose a difficulty as you can now have a
disparity error where the original paper would not have found one.
Also as the disparity output is to be calculated first on the basis of
the input 10 bits, would this also not result in a difference between
the output from a pure 8b/10b based on the paper and one that is done
based on the spec.
I was wondering if my interpretation is correct. I looked at the
original FC-PH standard (rev 4.3, June 1, 1994, I guess it is a pre-
release one) and it states that the definition is based on (and is in
basic agreement with) the Widmer/Franasak (not sure of the correct
spelling for the name) paper/patent.
The verbiage above seems to indicate there could be differences or it
could be to avoid issues with patent?
If it is to indicate differences, are there any other differences that
I might be missing.
Thanks for any input you have.
with regards
Arvind R.
P.S. On an unrelated note, has anyone had a chance to look at the
symbols used in the patent/paper. What does OR DOT mean - it is an OR
or NOR? From what I can see it seems to be OR.
| 
Yahoo! 
Google 
Windows Live 
del.icio.us 
digg 
Netscape 
XML site map