How To: Estimate System (transfer Function) from the Input and Output Data

Hell there.

Estimating the system characteristics (esp. the transfer function) by using the input and output data and number of poles and zeroes is pretty much straight forward in Matlab (using iddata and tfest functions). I am trying to achieve the same in Scilab. It would be really great if someone could guide me in achieving the same.

Scilab 09-09-20, 5:16 p.m. abbas14
I think I found a way. If I could geerate the frequency response from the input and output signals, I can use that frequency response to genetrate the transfer function through frep2tf. But the challenge now is to accurately generate the frequency response from the input and output signals. I even tried the time_id function, but it seems it isn't working. Throws the following error:
at line 30 of function time_id ( C:\Program Files\scilab-6.1.0\modules\cacsd\macros\time_id.sci line 42 ) With NaN or Inf a left division by scalar expected.
Hope someone would be able to suggest a solution...
11-09-20, 8:48 p.m. abbas14

Log-in to answer to this question.