Fixed-Point Designer provides data types and tools for developing fixed- and floating-point algorithms that are numerically efficient for embedded hardware deployment. It supports MATLAB® code, Simulink® models, and Stateflow® charts.
Optimizing numeric efficiency speeds up execution performance, decreases memory resources, and consumes less power. Full control is provided for all fixed-point properties including word length, scaling, and rounding. Double, single and half-precision floating-point types are supported. You can specify the data type attributes manually, or allow Fixed-Point Designer to automatically choose the data types for your design. Visualize the results of data type conversion with a log2 histogram displaying minimum and maximum data ranges and potential overflows or underflows. With Fixed-Point Designer, you can perform bit-true simulations to assess and optimize your fixed- or floating-point design versus the baseline design.
It provides a large number of functions such as math and bitwise operations, CORDIC implementations, and trig functions to support fixed-point computation. Fixed-Point Designer also optimizes memory efficiency with lookup tables that can be AUTOSAR compliant or HDL optimized. The cast and like functions let you separate MATLAB algorithms from their data type specifications to quickly explore design tradeoffs. Using Fixed-Point Designer with MathWorks® code generation products, you can generate pure integer C code or bit-true Verilog® and VHDL® code from your fixed-point design. There are also cycle-true HDL optimized blocks available.
