Block diagonal matrix of identity times scalar.

Hi,

I currently have a vector a = [1 2]', I wish to create a block diagonal matrix. Each block is identity (3x3) times the corresponding scalar in the vector a.

i.e. with a = [1 2]' I want to produce b = [1 0 0 0 0 0; 0 1 0 0 0 0; 0 0 1 0 0 0; 0 0 0 2 0 0; 0 0 0 0 2 0; 0 0 0 0 0 2]. The catch is a can be a vector of N x 1, thus b is of size 3*N x 3*N

The answer also has to be for loop free. I've tried using blkdiag() and eye() but have dimension issues in my multiplication

Answer 1

[Philippe Lebel]

here is my take.

a=[1,2];
c = round(a(1):1/(length(b(:,1))-1):a(2));
matrix = diag(c);

[Philippe Lebel]

never mind, it's way too specific, im working on something else

[Mohammed Kagalwala]

Thank you for your help!

[Philippe Lebel]

here it is:

clear
a=[1,2,3,4];
size_of_sub_matrices = 2;
a = arrayfun(@(x) ones(1,size_of_sub_matrices)*x, a, 'UniformOutput', false);
a = cell2mat(a);
matrix = diag(a)

matrix =

     1     0     0     0     0     0     0     0
     0     1     0     0     0     0     0     0
     0     0     2     0     0     0     0     0
     0     0     0     2     0     0     0     0
     0     0     0     0     3     0     0     0
     0     0     0     0     0     3     0     0
     0     0     0     0     0     0     4     0
     0     0     0     0     0     0     0     4