Timeline for Cost function for LTI system identification
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 23, 2019 at 19:03 | vote | accept | Jonas Schwarz | ||
| Apr 23, 2019 at 19:03 | comment | added | Jonas Schwarz | i think i understand it now! | |
| Apr 22, 2019 at 11:39 | comment | added | Matt L. | @JonasSchwarz: I've tried to explain the division by $A_{i-1}$ in the one but last paragraph of my answer. If convergence is achieved, the final cost function is equivalent to the one of the original non-linear problem given by Eqs $(1)$ and $(3)$ in my answer. | |
| Apr 22, 2019 at 10:24 | comment | added | Jonas Schwarz | Also, i made an error in my first question, i meant to ask about the integral over the difference of true and estimated transfer function instead of a frequency-dependent error. I'll edit the question. | |
| Apr 22, 2019 at 10:23 | comment | added | Jonas Schwarz | Thank you for your response! I now understand why the weighting is utilized. Could you elaborate on what is achieved by dividing by $A_{i-1}$ in the iterative cost function? Does it force convergence? | |
| Apr 22, 2019 at 9:43 | history | answered | Matt L. | CC BY-SA 4.0 |