Main Content

Fixed-Point Design

Floating-point to fixed-point conversion, fixed-point algorithm design

To model fixed-point signal processing algorithms and analyze the effects of quantization on system behavior and performance, you can use the DSP System Toolbox™ and Fixed-Point Designer™ functionality. For a list of DSP System Toolbox System objects that support fixed-point operations, see System Objects in DSP System Toolbox that Support Fixed-Point Design. For a list of blocks in DSP System Toolbox that support fixed-point operations, use the showsignalblockdatatypetable function.

Perform design tradeoff analyses and optimizations on these algorithms by simulating them with different word lengths, scaling, overflow handling, and rounding method choices before you commit to hardware. For an example that shows how to optimize fixed-point algorithms, see Optimized Fixed-Point FIR Filters.

You can design floating-point filters and convert the filters into fixed-point using the Fixed-Point Designer. This design workflow enables you to optimize the fixed-point filters and analyze the quantization effects. For more information, see Floating-Point to Fixed-Point Conversion.