How can I perform mathematical operations using integers in the Embedded MATLAB function block?

My model includes an Embedded MATLAB Function block, in which I attempt to perform mathematical operations using data of type integer (e.g., int8, uint8, int16, uint16, etc.). When I simulate my model, the following error is returned:

 Expected either a logical, char, single, or double.

 Found a int8.

 Function 'Embedded MATLAB Function' (#48.131.133), line 5, column 6:

 "x1"

Answer 1

[MathWorks Support Team]

This enhancement has been incorporated in Release 14 Service Pack 3 (R14SP3). For previous product releases, read below for any possible workarounds:

The ability to perform mathematical operations using integers in an Embedded MATLAB function block is not available in Simulink. This is stated as a caveat in the documentation for the Embedded MATLAB function block under the "Data Type Support" section, available by typing the following at the MATLAB 7.0 (R14) prompt:

web([docroot '/toolbox/simulink/slref/embeddedmatlabfunction.html'])

To work around this issue, perform mathematical operations using data of type double or single; afterwards convert the result to an integer, as accomplished with the following code:

y = int8((single(x1)+single(x2))/single(4));