WienIdentForDifNonsPolynomKernel

WienIdentForDifNonsPolynomKernel - % ............ KERNEL IS

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: % ............ KERNEL IS POLYNOMIAL................. % Identification of Wiener model. various breakpoints and slopes are % considered for nonlinear function. The signal used to excite the system % is not a small one. So even if the nonlinear function is invertible the % system is not identified correctly. Besides there is no noise in the % output. Roles of inputs and outputs changed %% -------- Try it for various nonlinearities. ---------------------------- clear all c u=normrnd(0,2,1,400); % A white gaussian input sequence u with length %400 0 mean and standard deviation 2 ut=normrnd(0,2,1,200); %input for testing. % e=normrnd(0,.2,1,400); % A white gaussian with zero mean and standart de %viation .2 with length 400. it is error term zt =zeros(1,400); a = [2.09 -2.063 1.209 -.4656 .1164 -.02975] ; % ai s b = [1 .8 .3 .4] ; N=200; r=7; m=6; % sg = 5; % bi s % now we will get the input output data. The last 200 datapoints will be % used for training % for t = 1:400 if (t==1) z(1,t) = b(1,t)*u(1,1); e = z(1,t); if (-20<e && e<20) y(1,t) = 1*(z(1,t)); % following was before: sinc(z(1,t))*z(1,t)^2 ;% no +e(1,t); % b(1,t)*sinc(u(1,1))*(u(1,1)^2) + e(1,t); end if (e<-20) y(1,t)= .5*(z(1,t)) -10 ; end...
View Full Document

Page1 / 4

WienIdentForDifNonsPolynomKernel - % ............ KERNEL IS

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online