I have created code for banded gauss elimination method but on running it gives an error of index exceeds matrix dimensions

function x = gauss_elimination(K,M);

ln = length(M); x = zeros(ln,1);
for n=1:ln-1

        for i=n+1:ln
             m = K(i,n)/K(n,n);
             for j=n+1:ln
               K(i,j) = K(i,j)-m*K(n,j);
             end
             M(i) = M(i)-m*M(n);
        end
  end
  x(ln) = M(ln)/K(ln,ln);
  for i=ln-1:-1:1
     S = M(i);
     for j=i+1:ln
       S = S-K(i,j)*x(j);
     end
     x(i) = S/K(i,i);
  end
  end
  %%another function created to call gauss elimination function in it
  clc
  clear 

%Two additional matrices A and B were added for checking the gauss elimination
%function:

A= [ 170.4105 37.9473 -113.8420 -37.9473;
    37.9473 69.2176 -37.9473 -12.6491;
    -113.8420 -37.9473 120.9131 45.0184;
    -37.9473 -12.6491 45.0184 19.7202];

B = [0; 0; 0; -1000];

K = xlsread('HW05_Coef.xlsx');
M = csvread('HW05_Const.csv');

gauss_elimination(K,M)

[Walter Roberson]

You would get that error if you confused your K and your M when you defined them. For example if you called with (B,A) instead of (A,B)

[Martin Olafsen]

What's the error message, doesn't it say which line has an issue? Do you check that the matrices are of valid size? For me the function ran using the A and B matrices(though I don't know what gauss elimination should do).

[Haseeb Ahmed Janjua]

it gives me the error of index exceeds matrix dimensions when im reading matrix from csv files And my K in csv file is 1092x1 and M is 1092x46

[Walter Roberson]

Your code assumes that K is square and that M is a vector.

Answer 1

[Martin Olafsen]

You use the length of M to iterate through the matrices, even though the amount of rows is different from the amount of columns in K.