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

RE: Optical Simulation: FOCSS


I am not implying any recommendation.

It is obvious to me that this model is critical to the development of the
10GbE standard. If there is additional information that can be used to
improve, correct, correlate,  or even replace the model we are using, it is
in our interest to understand this.

802.3 has never been timid about utilizing other technology to improve its
efforts. There is no reason this shouldn't extend to our optical link
modeling tools. 

So, is there anything here for us or no?


>-----Original Message-----
>From: petarp@xxxxxxxxxx [mailto:petarp@xxxxxxxxxx]
>Sent: Wednesday, July 05, 2000 1:07 PM
>To: ghiasi; stds-802-3-hssg@xxxxxxxx;
>Jonathan.Thatcher@xxxxxxxxxxxxxxxxxxxx; dolfi@xxxxxxxxxxxxxxxxxx
>Subject: Re: Optical Simulation: FOCSS
>Hi all,
>I evaluated several packages during the last year. Some of them were
>mentioned before, some not. In general, all the simulators 
>work on single
>mode fiber (is anyone surprised?).  All of them have the basic rate
>equations laser model (which, BTW is not exactly correct for 
>DFB lasers).
>Some offer other laser models (Transmission Line Laser Model 
>etc.). Most of
>them include noise driven laser rate equations, but they use the same
>numerical approach as for the Rate eq.  without noise.
>Fiber models are limited to single mode (but the models are 
>fairly complex,
>including many effects). One company offers Gaussian model for the MMF
>transfer function.
>The receivers are PIN or APDs, and additional blocks 
>representing preamps
>and amplifiers can be added, but without integrated approach.
>WDM systems can be simulated in all of them (they are 
>primarily written for
>simulating WDM links)
>Most of these products interface with other programs (notably 
>Matlab) and
>also offer the user to write his own models (beware, some do not!), and
>require substantial amount of training to use their full 
>capabilities. With
>the price rang of 10K-35K, I wonder how many are willing to pay.
>Jonathan, how do you envision the use of these tools? How many copies
>should we have, who should use them? Should we also take advantage to
>further refine the parameters for the other PMDs?
>For those of you interested, here are some names and websites, 
>the ones already mentioned (Optsim and IPSIS):
>Virtual Photonics  - PTDS -
>RSOFT  - LinkSim -
>ZKOM - Photoss -
>Artis - OPTSIM - www.artis-softwarew
>Ipsis - COMSIS -
>While there are some other software packages, I view these as the most
>serious ones.
>Petar Pepeljugoski
>IBM Research
>P.O.Box 218
>Yorktown Heights, NY 10598
>e-mail: petarp@xxxxxxxxxxxxxx
>phone: (914)-945-3761
>fax:        (914)-945-4134
>ghiasi <Ali.Ghiasi@xxxxxxxxxxx> on 07/05/2000 02:22:21 PM
>Please respond to ghiasi <Ali.Ghiasi@xxxxxxxxxxx>
>Sent by:  owner-stds-802-3-hssg@xxxxxxxx
>To:   stds-802-3-hssg@xxxxxxxx, Jonathan.Thatcher@xxxxxxxxxxxxxxxxxxxx,
>      dolfi@xxxxxxxxxxxxxxxxxx
>cc:   Ali.Ghiasi@xxxxxxxxxxx
>Subject:  Re: Optical Simulation: FOCSS
>Jonathan and Dave
>I have come across software called COMSIS 8 at, 
>which deals
>with laser chirp and non linearities as well.  Hopefully you know some
>Optisim made by has CW Lorenztzian 
>laser source and
>rate equation laser source.
>The above software only handles single mode lasers and fibers.
>Good Luck,
>Ali Ghiasi
>Sun Microsystems
>> Date: Mon, 3 Jul 2000 13:48:29 -0700 (PDT)
>> From: Dave Dolfi 3764 <dolfi@xxxxxxxxxxxxxxxxxx>
>> To: stds-802-3-hssg@xxxxxxxx, Jonathan.Thatcher@xxxxxxxxxxxxxxxxxxxx
>> Subject: Re: Optical Simulation: FOCSS
>> Mime-Version: 1.0
>> Content-Transfer-Encoding: 7bit
>> Content-MD5: eBETxSgF8nay89A9o3Pk4w==
>> X-Resent-To: Multiple Recipients <stds-802-3-hssg@xxxxxxxxxxxxxxxxxx>
>> X-Listname: stds-802-3-hssg
>> X-Info: [Un]Subscribe requests to  majordomo@xxxxxxxxxxxxxxxxxx
>> X-Moderator-Address: stds-802-3-hssg-approval@xxxxxxxxxxxxxxxxxx
>> Jonathan,
>> I looked at the web site and it seems to be a typical
>> numerical simulator with a couplie of exceptions.  It
>> seems to be set up for multilevel modulation and it
>> seems to have some sort of noise model, since RIN
>> is one of the input parameters.  However, since the
>> linewidth enhancement factor is NOT listed as an
>> input parameter, it may not deal with chirp.  There
>> are several modulator options so I'm guessing that it
>> is meant to model a system in which the laser is running
>> cw and the modulator is external.
>> Dave Dolfi
>> Agilent Technologies