The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> Measuring Jitter from an Eye-Diagram
https://designers-guide.org/forum/YaBB.pl?num=1155094031

Message started by always@smart on Aug 8^th, 2006, 8:27pm

Title: Measuring Jitter from an Eye-Diagram
Post by always@smart on Aug 8^th, 2006, 8:27pm

Dear All,

I have a query here.

Does anyone here know how to use the calculator or other way to get the jitter from an eye-diagram without DIY.

THank in advance.

smart

Title: Re: Measuring Jitter from an Eye-Diagram
Post by sugar on Aug 9^th, 2006, 10:44pm

use verilog-a

Title: Re: Measuring Jitter from an Eye-Diagram
Post by always@smart on Aug 10^th, 2006, 8:40pm

Hi sugar,

Thank you for taking initiative to reply me.

Do you hav any example for it by using verilog-a?

regards,
smart

Title: Re: Measuring Jitter from an Eye-Diagram
Post by always@smart on Aug 14^th, 2006, 2:12am

can anyone here please help me on this issue?

Thanks

Regards,
Smart

Title: Re: Measuring Jitter from an Eye-Diagram
Post by Ken Kundert on Aug 14^th, 2006, 10:09am

If you are using the Cadence software, you might want to look around in the libraries provided with Artist. I seem to remember there being a verilog-a model that was designed to produce eye-diagrams.

-Ken

Title: Re: Measuring Jitter from an Eye-Diagram
Post by Jess Chen on Aug 14^th, 2006, 10:37am

Look in rfLib. The catatory is "measurement". The element is called eye_diagram_generator. It is documented in the SpectreRF User's Guide. There is also a new function in the waveform calculator for creating eye-diagrams. The rfLib element does not compute jitter directly and I don' know if the waveform calculator does it. However, as "sugar" suggests, you should be able to compute jitter using VerilogA. I would measure jitter with the zero-crossings. You can use the @cross statement to detect the zero crossing. You may get some ideas from the rfLib/comms_instr element for creating histograms and statistics from the zero-crossings using VerilgoA You will have to subtract off k*T from the zero crossing where T is the symbol period and k is the kth crossing. But you may be able to use a modulo function instead. You should be able to compute all the statistics on the fly, i.e. recursively, instead post-processing the data. Again, the comms_instr block may give you some ideas on how to do that. If you get it working, I am sure most of us would like to see your code posted to this forum.

Another option is to write the zero crossings to a file using VerilogA and then post-process with Matlab, Mathematica, or Excel.

Title: Re: Measuring Jitter from an Eye-Diagram
Post by always@smart on Aug 14^th, 2006, 7:48pm

Dear Ken Kundert and Jess Chen,

Thank you for the guide.

Yes, I'm using Cadence. I'm not familiar with the verilogA, so i wonder if that is a way to use the Ocean built in function with some mathematical formula to obtain the jitter value?

or someone can help me in verilogA (to get the jitter)?

Please shed me more light ;)

Regards
Smart

Title: Re: Measuring Jitter from an Eye-Diagram
Post by jbdavid on Aug 30^th, 2006, 11:12pm

I seem to recall a Jitter measurement block in bmslib as well
(search the forum for bmslib- its been discussed before..)
Jonathan