Error??"Subscript indices must either be real positive integers or logicals."

Error in decalage (line 6) z(1:s-1,1:s-1)=y((p-s+2):p,(p-s+2):p);

Error in reconst_mat (line 51) mrx=decalage(mrx,g1);

%------------------function---------
function [mrx]=reconst_mat(zz1,zz2,zz3,zz4,filter)

%zz1=zz1*2;zz2=zz2*2;zz3=zz3*2;zz4=zz4*2;

       [hh1,hh2,gg1,gg2] = wfilters(filter);
       [d1,d]=size(hh1);
        for i=1:d
         h1(i)=hh1(d-i+1);
         h2(i)=hh2(d-i+1);
         g1(i)=gg1(d-i+1);
         g2(i)=gg2(d-i+1);
      end

g2 = logical(g2);
g1(g2)

[p1,p2]=size(zz1);
[s1,s2]=size(g1);

k=1;
for i=1:p1
   z1(k,:)=zz1(i,:);z1(k+1,:)=0;
   z2(k,:)=zz2(i,:);z2(k+1,:)=0;
   z3(k,:)=zz3(i,:);z3(k+1,:)=0;
   z4(k,:)=zz4(i,:);z4(k+1,:)=0;
   k=k+2;
end

m1=mat1(p1*2,p2*2,g1);
m2=mat1(p1*2,p2*2,g2);

y1=m1*z1;  y2=m2*z2;  y3=m1*z3;  y4=m2*z4;

yy1=y1+y2;
yy2=y3+y4;
clear y1 y2 y3 y4
k=1;
for i=1:p2
   x1(:,k)=yy1(:,i);x1(:,k+1)=0;
   x2(:,k)=yy2(:,i);x2(:,k+1)=0;
   k=k+2;
end
clear m1 m2

m1=mat2(p1*2,p2*2,g1);
m2=mat2(p1*2,p2*2,g2);

y1=x1*m1;
y2=x2*m2;
mrx=y1+y2;
mrx=decalage(mrx,g1);

%-----------------function 2-------------------------

function [z]=decalage(y,h)

[s1,s]=size(h); %s1=1, s=10
[p,p]=size(y);% p=4*4

z(1:s-1,1:s-1)=y((p-s+2):p,(p-s+2):p);

z(s:p,s:p)=y(1:(p-s+1),1:(p-s+1));
z(s:p,1:(s-1))=y(1:p-s+1,p-s+2:p);
z(1:s-1,s:p)=y(p-s+2:p,1:p-s+1);

%---------values in this code are as follow-------------------

mrx =

   1.0e+03 *

    3.5800    1.8740    3.5809    1.9055
    2.0416    0.9281    2.0859    1.2452
    3.5428    1.9043    3.5232    1.8226
    1.7251    1.0687    1.6663    0.7227

g1 =

Columns 1 through 5

         0         0   -0.0645   -0.0407    0.4181

Columns 6 through 10

    0.7885    0.4181   -0.0407   -0.0645         0

y1 =

 -162.1863   18.5793 -161.3208   18.6522
 -111.2043    8.3814  -66.9020   12.1164
 -147.8071   18.9622 -167.3815   17.3119
  -48.3556   11.4247 -107.1290    6.4696
y2 =

   1.0e+03 *

    3.7422    1.8554    3.7422    1.8868
    2.1528    0.9197    2.1528    1.2331
    3.6906    1.8853    3.6906    1.8052
    1.7735    1.0572    1.7735    0.7162

m1 =

   -0.0645   -0.0407    0.4181         0
         0   -0.0645   -0.0407    0.4181
    0.4181         0   -0.0645   -0.0407
   -0.0407    0.4181         0   -0.0645
m2 =

     1     1     1     0
     0     1     1     1
     1     0     1     1
     1     1     0     1
y =

   1.0e+03 *

    3.5800    1.8740    3.5809    1.9055
    2.0416    0.9281    2.0859    1.2452
    3.5428    1.9043    3.5232    1.8226
    1.7251    1.0687    1.6663    0.7227
h =

Columns 1 through 5

         0         0   -0.0645   -0.0407    0.4181

Columns 6 through 10

    0.7885    0.4181   -0.0407   -0.0645         0

s=10 p=4

 please help me to solve the error

[Guillaume]

I find it very difficult to read code of such length where the variables are just meaningless one letter characters. I find I constantly have to go back to where the variable has been declared to remember what it is.

I strongly advise you to use longer variable names that have meaning. You may know now what z, s, p, h, etc. stand for, but if you ever have to go back to this code in the future, you'll have a nightmare figuring it out.

Comments in the code would also help greatly.

[Guillaume]

[p,p]=size(y);

You're assigning the number of columns and the number of rows to the same variable p. After that line p will always be the number of columns regardless of the number of rows. Your y matrix is probably square so it does not matter, but to me that screams of a bug waiting to happen.

Answer 1

[Guillaume]

z(1:s-1,1:s-1)=y((p-s+2):p,(p-s+2):p);

For that line to work s must be at least two (so that s-1 >= 1) and p must be at least 2 greater than s (so that p-s+2 >= 1).

According to you p = 4 and s=10, so p-s+2 == -4. That's not a valid index.