| Products & Services | Solutions | Academia | Support | User Community | Company |
| Download Product Updates | | | Get Pricing | | | Trial Software |
| Documentation → Filter Design Toolbox |
| Contents | Index |
| Learn more about Filter Design Toolbox |
This table summarizes what's new and changed in Version 4.0 (R2006b):
New Features and Changes | Version Compatibility Considerations | Fixed Bugs and Known Problems | Related Documentation at Web Site |
|---|---|---|---|
Yes | No | Printable Release Notes: PDF |
New features and changes introduced in this version are described here.
Blocks for Designing Filters Added to Signal Processing Blockset (Filter Design Toolbox Required)
Support for Automatically Converting Floating-Point Filters to Fixed-Point Based on Input Data
This release adds a block library to the Signal Processing Blockset product. The new library contains blocks that design single- and multirate filters using the new filterbuilder filter design dialog boxes. If you have a license for Filter Design Toolbox, you can use the new blocks to design and implement filters in simulations. Users who do not have Fixed-Point Toolbox licenses can run models that contain the new blocks, but they cannot change the filter designs in the blocks.
Two new aspects of analysis allow you to convert filters from floating-point to fixed-point format automatically:
Scaled doubles logging
mfilt and dfilt objects now support the scaled double data type. Scaled doubles data types act like fixed-point data types, allowing you to work with both fixed-point and scaled doubles in the same calculation. Allowing this overcomes the limitation that math is permitted only between the same data types. The value ScaledDouble has been added to the DataType property of the numerictype object. The following values have also been added to the DataTypeMode property of the numerictype object:
Scaled double: binary point scaling
Scaled double: slope and bias scaling
Scaled double: unspecified scaling
A new method, autoscale, automatically converts an input filter from floating-point format to fixed-point format based on the results of filtering a set of data. autoscale attempts to scale the filter to prevent overflows in all filter computations.
The combination of these new capabilities provides the conversion support.
A new function, filterbuilder, provides a graphical tool for designing single- and multirate filters in both floating-point and fixed-point forms. For more information, refer to filterbuilder.
We upgraded the Farrow filters to provide fixed-point support. Similar to the fixed point for all other dfilt objects.
Add the new method fdesign.fracdelay for designing filters that offer fractional delays. To support designing filters with fdesign.fracdelay, we added a lagrange design method.
Use this new method to estimate the frequency response of a dfilt or mfilt object. freqrespest uses filtering to estimate the filter response. To support this new method, we added freqrespopts, an object that contains the parameters for freqrespest.
With the addition of this feature, you can create parallel structures composed of multirate filters. This extends the parallel filter capability using dfilt.parallel by letting you use multirate filters as the input filters. To create parallel filter objects from two or more multirate filters, the individual filters must have the same rate change factors.
All specification objects now support coupled-allpass filter designs as structures.
![]() | Version 4.1 (R2007a) Filter Design Toolbox | Version 3.4 (R2006a) Filter Design Toolbox | ![]() |

Learn how to apply early verification to your development process through these technical resources.
How much time do you spend on testing to ensure implementation meets system-level requirements?
| © 1984-2009- The MathWorks, Inc. - Site Help - Patents - Trademarks - Privacy Policy - Preventing Piracy - RSS |