Deciding on Data Representation and Storage
After you validate a model and produce a specification, you
design the software by using a system simulator or rapid prototyping
hardware.
The Real-Time Workshop and Real-Time Workshop Embedded Coder products
support the Simulink built-in data types. In addition, you can
use the Simulink Fixed Point product
to specify fixed-point math for Simulink, Stateflow, and Embedded MATLAB models.
Fixed-point processing is especially important in mass production
environments that cannot afford microprocessors with floating-point
units. The process of converting from floating-point to fixed-point
math is time consuming, whether or not you are using models. The Simulink Fixed Point product
includes conversion and scaling tools that help automate that process.
The Simulink Fixed Point product offers bit-wise accurate
simulation results when compared with the behavior on the actual embedded
microprocessor. It does this by using integer word sizes and other
hardware characteristics that you specify when you configure the model.
You do not need to generate code to perform fixed-point simulation
or analysis. However, after assessing the fixed-point model and confirming
that the results match those of the specification, or other reference,
you can generate code for implementation and further analysis. The Real-Time Workshop Embedded Coder product
has specific features that can help you generate highly efficient
fixed-point code. For example, you can explicitly suppress the generation
of floating-point code.
 | Defining Data Representation and Storage for
Code Generation | | Parameter Considerations |  |
Learn more about Simulink through this collection of videos, articles, technical literature and the Getting Started with Simulink Guide.
Get this Simulink Kit
Store 
