Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: Issues concerning 10GbE speed standards

Hi Rich,

> Kamran,
> I agree completely with your assertions and conclusions about the benefits of a
> well designed FEC such as Reed-Solomon.
> In addition, I believe that other coding requirements such as special symbol
> support, transition density, DC-balance, error containment (no multiplication of
> bit errors) may be combined with an FEC requirement and result in a code with
> fairly low overhead.

I agree. It would be great to incorporate all of the above at low cost, 
and have the further possibility of performing FEC (decoding of parity bits 
is clearly optional), by just having a provision for it (allocating bits for 
FEC purpose, and defining the FEC code so that we don't run into the problem
of proliferation of proprietary FEC codes).