how to convert file.m into block in Simulink ?

hi,

I wrote a function that produce 40 binary bits vector according to another 20 bits vector. In Matlab it works ok.

I want to convert it into a blockqmodel in Simulink.

Anybody knows how to do it ?

for example:

function gate_h = gate_array(I_H, indx_h)  
no_cells = 20;    
if(I_H > 0) 
  TMP_reg = [zeros(1, no_cells-indx_h), ones(1, indx_h)]; 
else 
  TMP_reg = [ones(1, indx_h), zeros(1, no_cells-indx_h)];
end  
bitpatterns = {[0 1], [1 0]};
gate_h = cell2mat(bitpatterns(TMP_reg + 1));

Thanks, Henry

Answer 1

[Walter Roberson]

Try using the following as a MATLAB Function block

function gate_h = gate_array(I_H, indx_h)
no_cells = 20;
gate_h = zeros(1, no_cells * 2);
gate_h(1:2:indx_h*2) = 1;
gate_h(indx_h*2+1:2:no_cells*2-1) = 1;
if I_H > 0
  gate_h = 1 - gate_h;
end

[Henry Buck]

great,

no error at all...looks ok. need to check the simulation results.

What about when I_H<0 ?

Thanks, Henry

[Walter Roberson]

I miscalculated mentally.

 function gate_h = gate_array(I_H, indx_h)
  no_cells = 20;
  gate_h = zeros(1, no_cells * 2);

if I_H > 0
  gate_h(2:2:(no_cells-indx_h)*2) = 1;
  gate_h((no_cells-indx_h)*2+1:2:no_cells*2-1) = 1;
else
  gate_h(1:2:indx_h*2-1) = 1;
  gate_h((indx_h+1)*2:2:no_cells*2) = 1;
end

[Henry Buck]

Thanks, But something a bit different. for example: if TMO_reg above is TMP_reg = [0 0 1 1 0 0 0 1] % in that case no_cells = 8. I want that gate_h will cange to gate_h = [0 1 0 1 1 0 1 0 0 1 0 1 0 1 1 0]. that values of gate_h will change until new TMP_reg appear: TMP_reg = [1 1 0 0 0 1 0 0] then gate_h = [1 0 1 0 0 1 0 1 0 1 1 0 0 1 0 1]. the bits vector gate_h is dubble then vector TMP_reg.

Thanks, Henry

[Henry Buck]

Hi,

I made a simple change and it seems to be works ok.

%%%%%%%%%%%%

function gate_h = gate_array_h6(I_H, indx_h)

no_cells = 20;
gate_h = zeros(1, no_cells * 2);
gate_h(1:2:indx_h*2) = 0;
gate_h(indx_h*2+1:2:no_cells*2-1) = 0;

if I_H > 0
  gate_h(2:2:(no_cells-indx_h)*2) = 1;
  gate_h((no_cells-indx_h)*2+1:2:no_cells*2-1) = 1;
else
  gate_h(1:2:indx_h*2-1) = 1;
  gate_h((indx_h+1)*2:2:no_cells*2) = 1;
end

%%%%%%%%%%%%

No E R R O R at all ....:-)).

Thanks a lot,

Henry

[Henry Buck]

Hi,

For changing direction of the gate_h bits vector to complement gate_l bit vector, what do I need to change ? for example: in one block I got gate_h= [0 1 0 1 0 1 0 1 1 0 1 0]. and in anothe block I want to get gate_l = [1 0 1 0 1 0 1 0 0 1 0 1].

its like gate_l is complement of gate_h

Answer 2

[Rick Rosson]

Have you tried using the MATLAB Function block? It's in the User Defined Functions sub-library in base Simulink.

[Henry Buck]

Yes I did, But when I add that code into MATLAB Function code, after running I got lots of error relate to the code above.

I dont know how solve it.

Thanks, Henry

[Henry Buck]

Hi, I made a simple change and it seems to be works ok.

%%%%%%%%%%%% function gate_h = gate_array_h6(I_H, indx_h)

no_cells = 20; gate_h = zeros(1, no_cells * 2); gate_h(1:2:indx_h*2) = 0; gate_h(indx_h*2+1:2:no_cells*2-1) = 0;

if I_H > 0 gate_h(2:2:(no_cells-indx_h)*2) = 1; gate_h((no_cells-indx_h)*2+1:2:no_cells*2-1) = 1; else gate_h(1:2:indx_h*2-1) = 1; gate_h((indx_h+1)*2:2:no_cells*2) = 1; end %%%%%%%%%%%%

No E R R O R at all ....:-)).

Thanks a lot, Henry