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

RE: Issues concerning 10GbE speed standards

Hi Ed,

> Kamran:
> Thanks for very elaborated explanation related to FEC with which I am not
> quite familiar.
> From you description, quote: " FEC codes are meant for error correction
> only"; therefore, FEC is one of the error correction scheme.
> Usually, the bit error rate is defined without error corrections.  The
> purpose of BER is to measure the reliability of the over-all link, which
> should not subtract the errors corrected by the error-correction-codes.

Why is this a requirement to achieve BER without FEC ? Do you mean this a 
common practice among vendors or do you mean this is specified in some 
WAN standard ? 
PHY standards define a BER. A PHY standard should not specify by what means 
vendor A or B achieve that spec. If vendor A decides to meet the BER 
objective by using error correction, that should be allowed by the standard. 
Now vendor B may prefer not to use FEC, and achieve BER without FEC, that's 
fine too. It's probably a good thing to have provision for FEC.

I would like to add that when you decode parity bits (if you decide to
do so), you get directly a measurement of the input BER or link quality 
(before error correction). If this is an important parameter for WAN 
applications, an estimated link quality could be made available at the 
output of the decoder at no extra cost. If I understand well your
comment, you are suggesting that link quality (without error correction)
should be made available to the user. That's possible.

> However, users can chose to correct the errors -- one error, or multiple
> errors-- based on the reliability measure, BER, supplied by vendors and
> requirement of the application to implement error corrections.  Error
> correction is not free, which will use valuable resources.  For very
> unreliable tapes or disks, error correction is a requirement, however for a
> well designed semiconductor memory, error correction is not required. 

I agree, error correction is not free, say it costs X kgates. It is up 
to the system designers then to decide, in which applications (if any)
the overall system cost can be decreased by using error correction.

> The reason I also elaborate the issue is that BER is measured without error
> corrections.  Error correction can not be used to claim the BER is improved.
> Otherwise, we will not have the universal referencing point in discussing
> the reliability of a link based on BER.

I think my above comment was probably addressing this comment. Error
correction does not prevent from having a measurement of link quality
before error correction. In many schemes you directly get the number 
of errors at the input of the FEC decoder, and this measurement can 
be made available to the user. Hope I understood and answered your