Why do I receive an error when I use the IMPULSE function in the System Identification Toolbox 5.0.2 (R13)?

Using the following code to calculate the impulse response:

a=[1 1 -274/800 -304/8000 7669/8000 ...

     11652/16000 -1420/32000 -6636/32000 ...

     -2646/64000];

b=[zeros(1,100) 1 -1 21/16 -1 5/16];

m=idarx(reshape(a,[1 1 9]),reshape(b,[1 1 105]));

u=idinput(1e5,'rgs');

e=randn(1e5,1);

y=sim(m,[u e]);

d=iddata(y,u);

I receive the following error:

 ??? Index exceeds matrix dimensions.

 Error in ==> C:\MATLAB6p5\toolbox\ident\ident\@idmodel\impulse.m

 On line 104  ==>         varargin = varargin(1:kpf-1,kpf+2:end);

 Error in ==> C:\MATLAB6p5\toolbox\ident\ident\@iddata\impulse.m

 On line 161  ==>    impulse(model,varargin{2:end});

Answer 1

[MathWorks Support Team]

This is a bug in the System Identification Toolbox Version 5.0.2 (R13) in the way that the IMPULSE function handles the input arguments. There is an error on Line 104 of the IMPULSE function in System Identification Toolbox. Line 104 should look exactly like Line 123.

To work around this issue, you can:

1) Type the following command at the MATLAB command window to bring up the editor:

edit C:\MATLAB6p5\toolbox\ident\ident\@idmodel\impulse.m

2) Copy Line 123, which reads

varargin = varargin([1:kpf-1,kpf+2:end]);

3) Delete Line 104, which reads

varargin = varargin(1:kpf-1,kpf+2:end);

4) Paste the line copied from Line 123 to Line 104.

5) Save the file.