I'm not sure I quite understand what you want, but a couple of things I see. First, row is a scalar giving the number of rows in z, and therefore h=row/9 is also just a scalar. In that case, g=std(h) is the standard deviation of a set made up of the single number h, which, by convention, equals 0.
I'm guessing you really want to compute standard deviations of sub-groups of the elements of x. Is x a vector or a matrix (i.e., is c=1 or >1)?
If x is a vector, and you want to compute standard deviations of each successive, non-overlapping group of 9 elements of x, then if the number of elements of x is an even multiple of 9 (i.e., if h=row/9 is an integer), you can do this easily via
x2 = reshape(x,9,h); % reshape the vector x into a matrix with 9 rows and h columns
g = std(x2,1); % compute the standard deviation of each column
If h is not an integer, it means your last frame will have less than 9 samples in it. You have to figure out how you want to deal with that. For example, you may want to compute standard deviations only for frames with 9 elements in them, in which case:
flh = floor(h); % round h downward
x2 = reshape(x(1:9*flh),9,flh); % reshape the first 9*flh elements of x into a 9-by-flh matrix
g = std(x2,1); % compute the standard deviation of each column
If x is a matrix, then you need to provide some more detail about exactly what you're computing the standard deviation of.
