Automated Fixed-Point Conversion in MATLAB

Conversion to fixed point using the Fixed-Point Converter app

The Fixed-Point Converter app enables you to convert floating-point MATLAB^® code to fixed-point MATLAB code.

Use fiaccel to convert floating-point MATLAB code to fixed-point MATLAB code at the command line.

Apps

(Not recommended) Convert MATLAB code to fixed point

Functions

`fiaccel`	Accelerate fixed-point code or convert floating-point MATLAB code to fixed-point MATLAB code
`coder.config`	Create configuration object for fixed-point or single-precision conversion
`coder.approximation`	Create function replacement configuration object
`coder.allowpcode`	Control code generation from P-code files
`coder.const`	Fold expressions into constants in generated code
`coder.extrinsic`	Declare function as extrinsic and execute it in MATLAB
`coder.float2fixed.skip`	Exclude functions from fixed-point conversion (Since R2024b)
`coder.inline`	Control inlining of current function in generated code
`coder.inlineCall`	Inline called function in generated code (Since R2024a)
`coder.load`	Load compile-time constants from MAT-file or ASCII file
`coder.newtype`	Create `coder.Type` object to represent type of an entry-point function input
`coder.nonInlineCall`	Prevent inlining of called function in generated code (Since R2024a)
`coder.nullcopy`	Declare uninitialized variables in generated code
`coder.resize`	Resize `coder.Type` object
`coder.target`	Determine if code generation target is specified target
`coder.typeof`	Create `coder.Type` object to represent the type of an entry-point function input
`coder.unroll`	Unroll `for`-loop by making a copy of the loop body for each loop iteration
`coder.varsize`	Resolve size mismatch errors and declare upper bounds

Classes

`coder.mexconfig`	Code acceleration configuration object for use with `fiaccel`
`coder.FixPtConfig`	Floating-point to fixed-point conversion configuration object
`coder.ArrayType`	Represent set of MATLAB arrays acceptable for input specification
`coder.Constant`	Specification of constant value for code generation
`coder.EnumType`	Represent set of MATLAB enumerations acceptable for input specification
`coder.FiType`	Represent set of MATLAB fixed-point arrays acceptable for input specification
`coder.PrimitiveType`	Represent set of logical, numeric, or character arrays acceptable for input specification
`coder.StructType`	Represent set of MATLAB structure arrays acceptable for input specification
`coder.Type`	Represent set of MATLAB values acceptable for input specification

Topics

Automated Fixed-Point Conversion Workflow

Decide Which Workflow Is Right for Your Application
Learn the benefits of each conversion workflow to help you decide which one best fits your needs.
Fixed-Point Conversion Workflows
Learn which fixed-point conversion method best matches your end goal and your level of fixed-point expertise.
Automated Fixed-Point Conversion
Using the Fixed-Point Converter app for automated fixed-point conversion.
Automated Fixed-Point Conversion Best Practices
Generate fixed-point code according to best practices.
Propose Data Types Based on Simulation Ranges Using the Fixed-Point Converter App
Propose fixed-point data types based on simulation ranges using the Fixed-Point Converter app.
Propose Data Types Based on Derived Ranges Using the Fixed-Point Converter App
Propose fixed-point data types based on static ranges using the Fixed-Point Converter app.
Convert Fixed-Point Conversion Project to MATLAB Scripts
Use project settings for fixed-point conversion at the command line.

Propose Fixed-Point Types

Specify Type Proposal Options
Specify options when proposing types for automated fixed-point conversion.
View and Modify Variable Information
View and modify variable information for fixed-point conversion.
Detect Unexecuted and Constant-Folded Code
Learn how the app detects code that is not executed or code that is constant folded during simulation of your test file.

Convert to Fixed Point

Generated Fixed-Point Code
Avoid issues with generated fixed-point code.
Replacing Functions Using Lookup Table Approximations
Generate lookup table approximations to replace custom functions or functions that are not supported for fixed point.
Replace the exp Function with a Lookup Table
Replace a function with a lookup table approximation in fixed-point code generated with the Fixed-Point Converter app.
Replace a Custom Function with a Lookup Table
Replace a custom function with a lookup table approximation function by using the Fixed-Point Converter app.

Verify Fixed-Point Conversion

Visualize Differences Between Floating-Point and Fixed-Point Results
Use a custom plot function to compare the behavior of the generated fixed-point code against the behavior of the original floating-point MATLAB code.
Enable Plotting Using the Simulation Data Inspector
Inspect and compare floating-point and fixed-point logged input and output data.
Custom Plot Functions
Visualize numerical differences during fixed-point conversion.
Detect Overflows
Detect overflows using the app.

Supported Language Features

MATLAB Language Features Supported for Automated Fixed-Point Conversion
Learn which language features are and are not supported for automated fixed-point conversion.
Convert Code Containing Global Variables to Fixed-Point
Convert MATLAB code using global variables to fixed-point using the app.
Convert Code Containing Structures to Fixed Point
Convert a MATLAB algorithm containing structures to fixed point using the Fixed-Point Converter app.
Generate Fixed-Point MATLAB Code for Multiple Entry-Point Functions
Generate fixed-point code for multiple entry-point functions using the Fixed-Point Converter app.
Fixed-Point Code for MATLAB Classes
Use supported constructs and coding style best practices for fixed-point conversion of MATLAB classes.
Convert Code Containing Global Data to Fixed Point
Convert MATLAB code using global variables to fixed-point programmatically.
Convert Identical Functions Called with Different Data Types
Convert a MATLAB algorithm containing specialized functions to fixed point using the Fixed-Point Converter app.
Convert dsp.FIRFilter Object to Fixed-Point Using the Fixed-Point Converter App
Convert a dsp.FIRFilter System object™, which filters a high-frequency sinusoid signal, to fixed-point using the Fixed-Point Converter app.
System Objects Supported by Fixed-Point Converter App
Use the Fixed-Point Converter app to automatically propose and apply data types for commonly used system objects.

Command Line Workflow

Propose Data Types Based on Simulation Ranges Using the fiaccel Function
Propose fixed-point data types based on simulation ranges using the fiaccel function.
Propose Data Types Based on Derived Ranges Using the fiaccel Function
Propose fixed-point data types based on static ranges using the fiaccel function.
Detect Overflows
Detect overflows at the command line.
Replace the exp Function with a Lookup Table
Replace a function with a lookup table approximation in fixed-point code generated with the fiaccel function.
Replace a Custom Function with a Lookup Table
Replace a custom function with a lookup table approximation function by using the fiaccel function.
Enable Plotting Using the Simulation Data Inspector
Inspect and compare floating-point and fixed-point logged input and output data.
Visualize Differences Between Floating-Point and Fixed-Point Results
Use a custom plot function to compare the behavior of the generated fixed-point code against the behavior of the original floating-point MATLAB code.

Troubleshooting

Generated Fixed-Point Code

Avoid issues with generated fixed-point code.

Data Type Issues in Generated Code

Highlight potential data type issues in report.

Debug Numerical Issues in Fixed-Point Conversion Using Variable Logging

Learn how to debug your fixed-point code when you need more than out of the box conversion.

Why Does the Fixed-Point Converter App Not Propose Data Types for System Objects?

Troubleshoot missing data type proposals for System objects.

Slow Operations in the Fixed-Point Converter App

Determine whether code generation readiness screening is causing slow operations in the app.