[SI-LIST] Re: HSPICE - adding jitter to ethernet serial link
- From: "Alfred P. Neves" <al.neves@xxxxxxxxxxx>
- To: <Chris.Cheng@xxxxxxxxxxxx>
- Date: Wed, 15 Feb 2006 10:34:55 -0800
Chris,
Your right, adding RJ to an eye diagram running in HSPICE (or another
more Berkeley Spice type of simulator for that matter) does not add any
analysis capability due to the definition of RJ. Actually, the approach
can be misleading and does not lend to a complete methodology. A
bathtub curve generated by a BERT scan is still an extrapolation it
relates the TJ to the sample size (or BER) essentially, it is a much
better tool when adding the RJ contribution. Here is why:
Any guesses at eye closure or width is based on estimation of the impact
of convolving RJ and DJ contributions, and since RJ is unbounded, simply
running the simulation for a longer period of time will relate to a
larger TJ, or smaller eye width.
Take a Noisecom UFX7109 generator and sum AWGN into a limiting amp on
one input and a NRZ signal running on the other input so the resulting
output is high quality (long tails in the Gaussian pdf) RJ for the NRZ
data. Now, measure the jitter. The 1-sigma value will settle to a
number for a small number of samples, but the peak-peak jitter will
continue to increase with the sample rate. pm me for this data. I used
this to dispel the specification myth of peak-peak jitter from the clock
manufacturers a while back.
Examining individual trajectories due to ISI is relevant (Ransom
Stephens, one of our PhD guys, wrote a good paper on equalization at
DesignCon2006 on this), especially when the simulation relates
S-parameter characterization of the system.
We have developed a package model where we optionally include the DJ of
the device (TX), contributing DJ of the package, and S11 and S12. When
this package/device complete signal integrity model is included into an
interconnect model of the system we have a complete picture of how each
pathology in the system contributes to the total peak-peak DJ.
We deal with RJ differently, since it is different from the other
pathologies. RJ is a random noise-VCO origination versus passive
interconnect and limitations of the TX driver. Another ancillary
benefit of this method is that XTALK, which tricks many of the jitter
extraction methods since it looks like RJ (XTALK is deterministic jitter
since it is bounded) but is not, is completely extracted and analyzed.
Unfortunately, most of the tools and your eye viewing the eye diagram
cannot tell the difference between XTALK and RJ so we intentionally do
not analyze the RJ contribution when simulating eye diagrams for that
reason also.
Alfred P. Neves <*)))))><{
Teraspeed Consulting Group LLC
121 North River Drive
Narragansett, RI 02882
Hillsboro Office
735 SE 16th Ave.
Hillsboro, OR, 97123
(503) 679 2429 Voice
(503) 210 7727 Fax
Main office
(401) 284-1827 Business
(401) 284-1840 Fax
http://www.teraspeed.com
Teraspeed is the registered service mark
of Teraspeed Consulting Group LLC
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx]
On Behalf Of Chris Cheng
Sent: Wednesday, February 15, 2006 1:34 AM
Cc: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: HSPICE - adding jitter to ethernet serial link
May be its just me but Rj only makes sense when you are trying to
extract a BERT curve which seems to be a MATLAB job. If you have your Tx
Rj, channel model and jitter transfer of your CDR, the rest of the work
is straight forward.
>
>
> At 10:40 AM 2/7/2006 -0800, Ali Burney wrote:
>
>> Hi,
>>
>>
>> I am running gigabit Ethernet link simulation using HSPICE and have
>> a=20 question on adding jitter to the transmitted data. Is there any
>> way to=20 add Gaussian jitter to the transmitted data? I know that
>> tools such as
>>
>
>
>> Hyperlynx allow you to add transmit jitter.
>>
>>
>>
>> Thanks and best regards,
>>
>>
>>
>> A Burney
>>
>>
>>
>>
>>
>>
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List FAQ wiki page is located at:
http://si-list.org/wiki/wiki.pl?Si-List_FAQ
List technical documents are available at:
http://www.si-list.org
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List FAQ wiki page is located at:
http://si-list.org/wiki/wiki.pl?Si-List_FAQ
List technical documents are available at:
http://www.si-list.org
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
