Why My nurbs Function code not work?

Hi guys! Im studying the nurbs function and as first thing i looked for matlab code around of internet. I found some code but its not for me, too complicated and bugged! So i decided to write a matlab code for nurbs. Someone can help me?? This is my code:

function [C, N, U] = nurbs_function(n, t, w, P, U)

% n = curve degree
% t = knot vector
% w = weight vector
% P = control points
% U = evaluation poins [Optionals]

if nargin < 5
    U = linspace(t(n), t(end-n), 10*size(P,2));% allocate points uniformly,

elseif isscalar(U) && U > 1
    validateattributes(U, {'numeric'}, {'positive','integer','scalar'});
    U = linspace(t(d+1), t(end-d), U);  % allocate points uniformly
end

m = numel(U);
N = zeros(m, size(P, 2));
for j = 1 : m
    u = U(j);
    for i = 1 : numel(t) - n -1
        N(j, i) = bspline_basis(i, n, t, u); %here i calculate the basis function N in u
    end
end
C = zeros(2, m);
for j = 1 : m
    saved = [0;0];
    for i = 1 : size(N, 2)
        if N(j, i) ~= 0 
            temp = (N(j, i) * w(i) * P(:, i))/(N(j, i) * w(i)); %this is formula of nurbs, but doesn't work well
            saved = saved + temp;
        end
    end
    C(:, j) = saved;

end

end

And this is the bspline_basis function written by Levente Hunyadi:

function [y,x] = bspline_basis(j,n,t,x)
% B-spline basis function value B(j,n) at x.
%
% Input arguments:
% j:
%    interval index, 0 =< j < numel(t)-n
% n:
%    B-spline order (2 for linear, 3 for quadratic, etc.)
% t:
%    knot vector
% x (optional):
%    value where the basis function is to be evaluated
%
% Output arguments:
% y:
%    B-spline basis function value, nonzero for a knot span of n

% Copyright 2010 Levente Hunyadi

validateattributes(j, {'numeric'}, {'nonnegative','integer','scalar'});
validateattributes(n, {'numeric'}, {'positive','integer','scalar'});
validateattributes(t, {'numeric'}, {'real','vector'});
assert(all( t(2:end)-t(1:end-1) >= 0 ), ...
    'Knot vector values should be nondecreasing.');
if nargin < 4
    x = linspace(t(n), t(end-n+1), 100);  % allocate points uniformly
else
    validateattributes(x, {'numeric'}, {'real','vector'});
end
assert(0 <= j && j < numel(t)-n, ...
    'Invalid interval index j = %d, expected 0 =< j < %d (0 =< j < numel(t)-n).', j, numel(t)-n);

y = bspline_basis_recurrence(j,n,t,x);

function y = bspline_basis_recurrence(j,n,t,x)

y = zeros(size(x));
if n > 1
    b = bspline_basis(j,n-1,t,x);
    dn = x - t(j+1);
    dd = t(j+n) - t(j+1);
    if dd ~= 0  % indeterminate forms 0/0 are deemed to be zero
        y = y + b.*(dn./dd);
    end
    b = bspline_basis(j+1,n-1,t,x);
    dn = t(j+n+1) - x;
    dd = t(j+n+1) - t(j+1+1);
    if dd ~= 0
        y = y + b.*(dn./dd);
    end
elseif t(j+2) < t(end)  % treat last element of knot vector as a special case
  y(t(j+1) <= x & x < t(j+2)) = 1;
else
  y(t(j+1) <= x) = 1;
end

I suppose that this code is right. I tryed my code with this params:

t = [0 0 0 1 2 5 5.5 6 9 11 13 13 13];
w = [1.2 3 5 1 0 0.5 2 3 5];
P = [2 5 3 3 6 10 10 7 9; 0 1 4 8 10 8 4 1 0];
n = 3;

The result is this: a matrix 2 x 90 with this value:

    C(:, 1: 10):
       0     7     7     7     7     7     7     7     7    10
       0     1     1     1     1     1     1     1     1     5
    C(:, 41 : 51):
        11   NaN   NaN   NaN   NaN   NaN   NaN   NaN   NaN   NaN   NaN
        13   NaN   NaN   NaN   NaN   NaN   NaN   NaN   NaN   NaN   NaN
    C(:, 81 : 90):
        27    27    27    27    27    27    27    27    27    17
        13    13    13    13    13    13    13    13    13     5

This are only some portion of matrix... but as you can see the matrix is abnormal! Can someone help me? Where i wrong? Thanks

[Farid Freud]

j: % interval index, 0 =< j < numel(t)-n

how i can input j in matlab ?

as vector or point , in brackets or parentheses ?