Filter Design Toolbox Release Notes | ![]() ![]() |
New Transformations Pane Added to FDATool
A new Transformations pane in FDATool provides access to the new transformation functions. All filter transformations work from the command line as well. In addition, Version 2.2 modifies the existing filter transformations in FDATool--they use the same input and output arguments but with modified transformation techniques.
Replaces the Transformations menu option in earlier toolbox versions and expands the capability to transform filters.
nlm Filter Analysis Method in FDATool
The new release of Filter Design Toolbox adds the noise loading method (nlm
) to FDATool. While this analytical method was in the toolbox as a command line function, nlm
is now available in the GUI.
To run the noise loading method in FDATool:
New Functions for Calculating Quantizer Noise Statistics
Three new functions for determining the noise statistics of a quantizer now appear in the toolbox:
errmean
--return the mean of the quantization error resulting from quantizing a signal
errpdf
--calculate the probability density function (pdf) of the quantization error
errvar
--return the variance of the quantization error resulting from quantizing a signal
Enhanced Quantizer Optimization
For this release, a new Optimization... option replaces the scaling functions on the quantization panel. Optimization enables you to specify a larger range of optimization options for your quantized filters. When you click Optimization... on the quantizer panel, you see the following Quantized Optimizations dialog, shown with the default settings:
For more information about using the new optimization features, refer to Optimizing the Quantization Process For Your Filter in the online help.
Enable and Disable Quantizers in a Filter
In the revised Quantized Filter pane in FDATool, you find a new option--six check boxes that enable or disable the quantizers in a filter. By clearing or selecting a check box, you determine whether to apply the associated quantizer to the filter during the quantization process. For example, you can elect not to apply the Input quantizer by clearing the check box Convert input to. During quantization, the inputs will be passed on without modification or changes to their values.
Quantizer Status Functions Added
To help you determine various characteristics of quantizers, the toolbox includes three new functions:
isfixed
--test and return whether a quantizer is fixed point
isfloat
--test and return whether a quantizer is floating point
isnone
--determine and return whether a quantizer has quantization mode equal to none
![]() | New Frequency Transformation Functions Added | Enhanced Quantized Filter Structures | ![]() |