Accelerating the pace of engineering and science

Best practices for Converting MATLAB Code to Fixed Point using Fixed-Point Designer

Register to watch video

Jared Schuler, MathWorks
Harshita Bhurat, MathWorks
Tom Bryan, MathWorks

MathWorks fixed-point team demonstrates new features, functionality and best practices for converting MATLAB code to fixed point using Fixed-Point Designer.  The MathWorks Fixed-Point Designer helps you design and convert your algorithms to fixed point. Whether you are simply designing fixed-point algorithms in MATLAB or using Fixed-Point Designer in conjunction with one of MathWorks’ code generation products, these best practices will help you get from generic MATLAB code to an efficient fixed-point implementation.

This webinar covers best practices for preparing your code for conversion, efficiently converting your code to fixed point, and optimizing your results. You will learn tips and tricks for managing bit growth, getting data type proposals based on range information, and optimizing your code to meet your fixed-point constraints.

Highlights include:

  • Prepare code for instrumentation, fixed-point conversion and acceleration
  • Manage datatypes and control bit growth using subscripted assignment
  • Explore fixed-point design trade-offs by separating datatype definitions from the algorithmic code
  • Instrumentation for collecting simulation minimums and maximums
  • Data type proposals based on range analysis
  • Histogram logging for optimizing fixed-point datatypes

Product Focus

  • Fixed-Point Designer
  • MATLAB Coder
  • Embedded Coder

Recorded: 7 Nov 2013