Genesys "math language"

[Mark W. Ingalls]

Howdy,

I am learning about transforming frequency domain (S-parameter) data to time domain data using IFFT() in Genesys 2012.

I didn't know this back then, but there has evidently been something called "Math Language" incorporated in Genesys since 2008. The Help file for 2012 says,

Math Language, along with most of its built-in functions, was designed to be compatible with m-file script syntax.

Question 1: Does the above quote imply that "Math Language" is patterned after Matlab?
Question 2: Can anyone recommend a good reference for understanding "Math Language," and especially FFT() / IFFT() applications?

Thanks for reading, and please: Do comment!
Mark

 

[Mark W. Ingalls]

(Yeah, I know; I'm responding to my own dead thread. I don't think it means I'm crazy, though.)

Just in case anybody searches up this poor little orphaned thread, the answer to Q1 is, "Yep, evidently." I took a look at Matlab's online help documentation, and found plenty of similar syntax. So the answer to Q2 is "STFW for Matlab's online help doc." 8-O

I'm still wading through FFT / IFFT writen mat'l, but haven't found anything with just the right level of detail while also providing the answers I seek. Probably end up writing my own. Still would appreciate any suggestions, though.

 

[biff44]

I was trying to figure out just what the ifft function was. I measured S11 on a cable with an open, short, load at the end, and then used genesys to plot ifft(s11) of the s2p file. I did get a spike at the delay length of the termination, and did get a big spike amplitude with open or short and a small spike amplitude with the 50 ohm load.

But it was a spike, not a step. So what is it really computing? The impulse response? I was hoping it would display a step response.

 

[Mark W. Ingalls]

Well, I'll let you know more as I learn it, biff44.

I did the same thing with electric TLINEs, btw.

We're trying to use time delay for filter tuning, a la Agilent AN 1287-8.

take care...

 

[biff44]

I got this answer on the agilent forum. Apparently ifft(S11) is the impulse response:

"The ifft() is the inverse fast fourier transform. When you apply this to s-parameters you get the impulse response. However like all ifft() functions you will need to fix the time scale since it will most likely think it is in seconds and not the proper units based on the delta frequency, if I recall correctly. "