How can i normalise my features vector?

I want to test normalizing my features vector in 2 ways.

#1 the sum of the entire normalized matrix is 1 #2 the sum of each row is 1

my features vector is a 2d matrix extracted from an ecg i already tried normalising using bsxfun and zscore, the results weren't that great but zscore did much much better than bsxfun,

[Adam]

What is your question exactly? If you want to do #1 and #2 just sum the matrix and divide by that or sum it per row and divide by that vector using ./