(Not recommended) Apply specified expression to input
Fcn is not recommended. For more complicated expressions, use the MATLAB Function block. For less complicated expressions, consider replacing with blocks that model the same behavior.
Simulink / UserDefined Functions
The Fcn block applies the specified mathematical expression to its input. The expression can include one or more of these components:
u
— The input to the block. If u
is a vector, u(i)
represents the i
th
element of the vector; u(1)
or u
alone
represents the first element.
Numeric constants.
Arithmetic operators (+  *
/ ^).
Relational operators (== != > < >= <=
) —
The expression returns 1 if the relation is true; otherwise, it returns
0.
Logical operators (&&  !) — The expression returns 1 if the relation is true; otherwise, it returns 0.
Parentheses.
Mathematical functions — abs
,
acos
, asin
, atan
,
atan2
, ceil
, cos
,
cosh
, exp
, floor
,
hypot
, log
, log10
,
power
, rem
, sgn
(equivalent to sign
in MATLAB^{®}), sin
, sinh
,
sqrt
, tan
, and
tanh
.
The Fcn block does not support round
and
fix
. Use the Rounding Function block
to apply these rounding modes.
Workspace variables — Variable names that are not recognized in the
preceding list of items are passed to MATLAB for evaluation. Matrix or vector elements must be specifically
referenced (e.g., A(1,1)
instead of A
for
the first element in the matrix).
The Fcn block observes the following rules of operator precedence:
( )
^
+ 
(unary)
!
* /
+ 
> < <= >=
== !=
&&

The expression differs from a MATLAB expression in that the expression cannot perform matrix computations. Also, this block does not support the colon operator (:).
Block input can be a scalar or vector. The output is always a scalar. For vector
output, consider using the Math Function block. If a block input is
a vector and the function operates on input elements individually (for example, the
sin
function), the block operates on only the first vector
element.
You cannot tune the expression during simulation in Normal or Accelerator mode (see How Acceleration Modes Work), or in generated code. To implement tunable expressions, tune the expression outside the Fcn block. For example, use the Relational Operator block to evaluate the expression outside.
The Fcn block does not support custom storage classes. See Apply BuiltIn and Customized Storage Classes to Data Elements (Embedded Coder).
Data Types 

Direct Feedthrough 

Multidimensional Signals 

VariableSize Signals 

ZeroCrossing Detection 

Interpreted MATLAB Function  MATLAB Function  MATLAB System