Perform operation on matrix without for loop

So I have a 122x64x64 matrix x that has a grid of signals over time. First dim is time, second is x and third is y. Since the signals are all small signals with large DC offsets, I want to remove the offsets from the signals for all time values. The offsets for each signal are stored in a 64x64 matrix.

Here is the code I am currently using:

for a=1:64
    for b=1:64
        x(:,a,b) = x(:,a,b) - offset(a,b);
    end
end

Is there a faster way to do this without a for loop? Later I want to do other similar operations, so a non-loop method would be preferred.

Answer 1

[Roger Stafford]

 [m,n] = size(offset);
 x = bsxfun(@minus,x,reshape(offset,1,m,n));

Answer 2

[José-Luis]

EDIT

 permute(bsxfun(@minus,permute(x,[2 3 1]),offset),[3 1 2]);

[Shawn]

This returns a mismatched dimensions error.

[José-Luis]

My bad, forgot about non-singleton dimensions...

x = rand(128,64,64);
offset = rand(64);
tic
your_mat = permute(bsxfun(@minus,permute(x,[2 3 1]),offset),[3 1 2]);
toc
your_mat_1 = x;
for a=1:64
    for b=1:64
        your_mat_1(:,a,b) = x(:,a,b) - offset(a,b);
    end
end
toc
all(your_mat(:) == your_mat_1(:))