How about the following?
In this case, C{1}, C{2}, ..., C{151} are 200-by-1 matrices each, extracted from original 30194-by-1 matrix.
% Sample 30194-by-1 data
A = rand(30194,1);
% Padding with "nan"
n = 200;
N = numel(A);
A = [A;nan(ceil(N/n)*n-N,1)];
% Divide A into 200-by-1 segments
C = mat2cell(A,repelem(n,ceil(N/n)));
