What is wrong with this forward difference central space scheme?

The differential equation to be solved is the 1D heat equation Uxx-Ut=0 with the initial condition : u(x, 0) = g(x) = sin(pi/2)x for 0<= x< =2 and Boundary Conditions: u(0, t) = 0; u(2, t) = 0 for t > 0. I am getting a zero matrix.

The code is as follows:

function [ U ] = FTCS( M,N )

% Discretization Scheme % N points in space % M points in time

h = 2/(N-1); k = 1/(M-1); tau = k/h^2;

U = zeros(M,N); t = 0; x = 0;

if tau>0.5 'This may yield an unstable solution'; end

%initial condition for j=1:N U(1,j) =0; end

% Boundary Condition 1 for j=1:M U(j,1) = 0; end

% Boundary Condition 2 for j=1:M U(j,N) = 0; end

%internal points

for j=1:M-1 %outer loop for time for i=2:N-1 %inner loop for space U(j+1,i)=tau*U(j,i-1)+(1-2*tau)*U(j,i)+tau*U(j,i+1); end end

end

[Torsten]

What do you expect if you start with U=0 for t=0 and U=0 at both boundaries ?

Best wishes

Torsten.