Simulate neural network with a new signal

Hello everyone.

I'm new in NN stuff and I've got a problem.

I want to simulate a new signal in my net, but not using the command "SIM". I want use the product of matrices. So when i compare the final result using the funcion "SIM" and my script the numbers are diferents.

my neural network:

my scrip for simulate a new signal and get the output:

%#######################################################
 Iw = net.IW;  %the input-to-hidden layer weights
 weights_input = Iw{1,1}';
 Lw = net.Lw; %the hidden-to-output layer weights
 weights_output = Lw{2,1};
 b1  = net.b(1); %the input-to-hidden layer bias
 Bias_1 = b1{1,1}';
 b2 = net.b(2);  %the hidden-to-output layer bias
 Bias_2 = b2{1,1}';
 clear Iw Lw b1 b2
%#######################################################
x = 1; %neurons

 for i = 1:x %neurons
     M(:,i) = Signal.*weights_input(:,i);
 end
 
 for i = 1:x
    S(:,i) = sum(M(:,i)) + Bias_1(:,i);     
 end
 
 for i = 1:x
    Output_1(:,i) = tansig(S(:,i)); %####################
 end
 
 %######################################
 %Output layer
 
 for i = 1:1 %neurons
     M_1(:,i) = Output_1.*weights_output(i,:);
 end
 
for i = 1:1
    S_1(:,i) = sum(M_1(:,i)) + Bias_2(:,i);     
end
 
Output_Final = logsig(S_1');