Interactive filter design
filterBuilder(h)
filterBuilder('response
')
filterBuilder
starts a interactive tool for building filters. It
relies on the fdesign
object-object oriented filter design
paradigm, and is intended to reduce development time during the filter design process.
filterBuilder
uses a specification-centered approach to find
the best algorithm for the desired response.
Note
You must have the Signal Processing Toolbox™ installed to use fdesign
and
filterBuilder
. Some of the features described below may
be unavailable if your installation does not additionally include the
DSP System Toolbox™. You can verify the presence of both toolboxes by typing
ver
at the command prompt.
For more information on how to use filterBuilder
, see Filter Builder Design Process.
To use filterBuilder
, enter filterBuilder
at
the MATLAB® command line using one of three approaches:
Simply enter filterBuilder
. MATLAB opens a dialog for you to select a filter response type. After you
select a filter response type, filterBuilder
launches the
appropriate filter design dialog box.
Enter filterBuilder(h)
, where h is an existing filter
object. For example, if h
is a bandpass filter,
filterBuilder(h)
opens the bandpass filter design dialog
box. The h
object must have been created using
filterBuilder
or using
fdesign
.
Note
You must have the DSP System Toolbox software to create and import filter System objects.
Enter
filterBuilder('
to
replace response
')response
with a response method from the
following table. MATLAB opens a filter design dialog that corresponds to the specified
response.
Note
You must have the DSP System Toolbox software to implement a number of the filter designs listed in the following table. If you only have the Signal Processing Toolbox software, you can design a limited set of the following filter-response types.
Response Method | Description of Resulting Filter Design | Filter Object |
---|---|---|
arbgrpdelay | Arbitrary group delay filter design | fdesign.arbgrpdelay (DSP System Toolbox) |
arbmag | Arbitrary magnitude filter design | fdesign.arbmag (DSP System Toolbox)
|
arbmagnphase | Arbitrary response filter (magnitude and phase) | fdesign.arbmagnphase (DSP System Toolbox)
|
audioweighting | Audio weighting filter | fdesign.audioweighting (DSP System Toolbox) |
bandpass
or bp | Bandpass filter | fdesign.bandpass (DSP System Toolbox) |
bandstop
or bs | Bandstop filter | fdesign.bandstop (DSP System Toolbox) |
cic | CIC filter | fdesign.decimator(M,'cic',...) or
fdesign.interpolator(L,'cic',...) See fdesign.decimator (DSP System Toolbox)
and fdesign.interpolator (DSP System Toolbox) |
ciccomp | CIC compensator | fdesign.ciccomp (DSP System Toolbox) |
comb | Comb filter | fdesign.comb (DSP System Toolbox) |
diff | Differentiator filter | fdesign.differentiator (DSP System Toolbox) |
fracdelay | Fractional delay filter | fdesign.fracdelay (DSP System Toolbox) |
halfband
or hb | Halfband filter | fdesign.halfband (DSP System Toolbox) |
highpass
or hp | Highpass filter | fdesign.highpass (DSP System Toolbox) |
hilb | Hilbert filter | fdesign.hilbert (DSP System Toolbox) |
isinc,
isinclp, or isinchp | Inverse sinc lowpass or highpass filter | fdesign.isinclp (DSP System Toolbox) and
fdesign.isinchp (DSP System Toolbox) |
lowpass
or lp | Lowpass filter (default) | fdesign.lowpass (DSP System Toolbox) |
notch | Notch filter | fdesign.notch (DSP System Toolbox) |
nyquist | Nyquist filter | fdesign.nyquist (DSP System Toolbox) |
octave | Octave filter | fdesign.octave (DSP System Toolbox) |
parameq | Parametric equalizer filter | fdesign.parameq (DSP System Toolbox) |
peak | Peak filter | fdesign.peak (DSP System Toolbox) |
Note
Because they do not change the filter structure, the magnitude specifications and
design method are tunable when using filterBuilder
.
The main pane of Filter Builder varies depending on the filter response type, but the basic structure is the same. The following figure shows the basic layout of the dialog box.
As you choose the response for the filter, the available options and design parameters displayed in the dialog box change. This display allows you to focus only on parameters that make sense in the context of your filter design.
Every filter design dialog box includes the options displayed at the top of the dialog box, shown in the following figure.
Save variable as — When you click
Apply to apply your changes or
OK to close this dialog box,
filterBuilder
saves the current filter to your
MATLAB workspace as a filter object with the name you enter.
View Filter Response — Displays the magnitude response for the current filter specifications and design method by opening the Filter Visualization Tool (FVTool).
Note
The filterBuilder
dialog box includes an
Apply option. Each time you click
Apply, filterBuilder
writes
the modified filter to your MATLAB workspace. This modified filter has the
variable name you assign in Save variable as. To apply
changes without overwriting the variable in you workspace, change the
variable name in Save variable as before you click
Apply.
There are three tabs in the Filter Builder dialog box, containing three panes: Main, Data Types, and Code Generation. The first pane changes according to the filter being designed. The last two panes are the same for all filters. These panes are discussed in the following sections.
The second tab in the Filter Builder dialog box is shown in the following figure.
The Arithmetic drop down box allows the choice of
Double precision
, Single
precision
, or Fixed point
. Some of
these options may be unavailable depending on the filter parameters. The following
table describes these options.
Arithmetic List Entry | Effect on the Filter |
---|---|
Double precision | All filtering operations and coefficients use
double-precision, floating-point representations and math. When
you use filterBuilder to create a filter,
double precision is the default
value for the Arithmetic property. |
Single precision | All filtering operations and coefficients use single-precision floating-point representations and math. |
Fixed point | This entry applies selected default values, typically used on
many digital processors, for the properties in the fixed-point
filter. These properties include coefficient word lengths,
fraction lengths, and various operating modes. This setting
allows signed fixed data types only. Fixed-point filter design
with filterBuilder is available only when you
install Fixed-Point Designer™ software along with DSP System Toolbox software. |
The following figure shows the Data Types pane after you
select Fixed point
for Arithmetic
and set Filter internals to Specify
precision
. This figure shows the Data Types pane
for the case where the Use a System object to implement filter
check box is not selected in the Main pane.
When you select Use a System object to implement filter check box in the Main pane, the Data Types pane appears as below:
Not all parameters described in the following section apply to all filters. For example, FIR filters do not have the Section input and Section output parameters.
Specify the format the filter applies to data to be filtered. For
all cases, filterBuilder
implements filters
that use binary point scaling and signed input. You set the word
length and fraction length as needed.
Choose how you specify the word length and the fraction length of the filter numerator and denominator coefficients:
Specify word length
enables you
to enter the word length of the coefficients in bits. In
this mode, filterBuilder
automatically
sets the fraction length of the coefficients to the
binary-point only scaling that provides the best possible
precision for the value and word length of the
coefficients.
Binary point scaling
enables
you to enter the word length and the fraction length of the
coefficients in bits. If applicable, enter separate fraction
lengths for the numerator and denominator
coefficients.
The filter coefficients do not obey the Rounding
mode and Overflow mode
parameters that are available when you select
Specify precision
from the
Filter internals list. Coefficients are always saturated and
rounded to Nearest
.
Choose how you specify the word length and the fraction length of the fixed-point data type going into each section of an SOS filter. This parameter is visible only when the selected filter structure is IIR and SOS.
Binary point scaling
enables
you to enter the word and fraction lengths of the section
input in bits.
Specify word length
enables you
to enter the word lengths in bits.
Choose how you specify the word length and the fraction length of the fixed-point data type coming out of each section of an SOS filter. This parameter is visible only when the selected filter structure is IIR and SOS.
Binary point scaling
enables
you to enter the word and fraction lengths of the section
output in bits.
Specify word length
enables you
to enter the output word lengths in bits.
Contains the filter states before, during, and after filter
operations. States act as filter memory between filtering runs or
sessions. Use this parameter to specify how to designate the state
word and fraction lengths. This parameter is not visible for direct
form and direct form I filter structures because
filterBuilder
deduces the state directly from
the input format. States always use signed representation:
Binary point scaling
enables
you to enter the word length and the fraction length of the
accumulator in bits.
Specify precision
enables you
to enter the word length and fraction length in bits (if
available).
Determines how the filter handles the output of product operations. Choose from the following options:
Full precision
—
Maintain full precision in the result.
Keep LSB
— Keep the
least significant bit in the result when you need to shorten
the data words.
Specify Precision
—
Enables you to set the precision (the fraction length) used
by the output from the multiplies.
Specify how the fixed-point filter performs arithmetic operations within the filter. The affected filter portions are filter products, sums, states, and output. Select one of these options:
Full precision
—
Specifies that the filter maintains full precision in
all calculations for products, output, and in the
accumulator.
Specify precision
—
Set the word and fraction lengths applied to the results
of product operations, the filter output, and the
accumulator. Selecting this option enables the word and
fraction length controls.
Selecting this option directs the filter to use signed representations for the filter coefficients.
Sets the word length for the associated filter parameter in bits.
Sets the fraction length for the associate filter parameter in bits.
Use this parameter to specify how you would like to designate the accumulator word and fraction lengths.
Determines how the accumulator outputs stored values. Choose from the following options:
Full precision
—
Maintain full precision in the accumulator.
Keep MSB
— Keep the
most significant bit in the accumulator.
Keep LSB
— Keep the
least significant bit in the accumulator when you need
to shorten the data words.
Specify Precision
—
Enables you to set the precision (the fraction length)
used by the accumulator.
Sets the mode the filter uses to scale the output data after filtering. You have the following choices:
Avoid Overflow
— Set the
output data fraction length to avoid causing the data to
overflow. Avoid overflow
is
considered the conservative setting because it is
independent of the input data values and range.
Best Precision
— Set the
output data fraction length to maximize the precision in the
output data.
Specify Precision
— Set
the fraction length used by the filtered data.
Parameters in this group control how the filter rounds fixed-point values and how it treats values that overflow.
Sets the mode the filter uses to quantize numeric values when the values lie between representable values for the data format (word and fraction lengths).
ceil
— Round toward positive
infinity.
convergent
— Round to the
closest representable integer. Ties round to the nearest
even stored integer. This is the least biased of the methods
available in this software.
zero/fix
— Round toward
zero.
floor
— Round toward negative
infinity.
nearest
— Round toward nearest.
Ties round toward positive infinity.
round
— Round toward nearest.
Ties round toward negative infinity for negative numbers,
and toward positive infinity for positive numbers.
The choice you make affects everything except coefficient values and input data which always round. In most cases, products do not overflow—they maintain full precision.
Sets the mode the filter uses to respond to overflow conditions in fixed-point arithmetic. Choose from the following options:
Saturate
— Limit the output to
the largest positive or negative representable value.
Wrap
— Set overflowing values to
the nearest representable value using modular
arithmetic.
The choice you make affects everything except coefficient values and input data which always round. In most cases, products do not overflow—they maintain full precision.
Specifies whether to cast numeric data to the appropriate accumulator format before performing sum operations. Selecting Cast before sum ensures that the results of the affected sum operations match most closely the results found on most digital signal processors. Performing the cast operation before the summation adds one or two additional quantization operations that can add error sources to your filter results.
If you clear Cast before sum, the filter prevents the addends from being cast to the sum format before the addition operation. Choose this setting to get the most accurate results from summations without considering the hardware your filter might use. The input format referenced by Cast before sum depends on the filter structure you are using.
The effect of clearing or selecting Cast before sum is as follows:
Cleared — Configures filter summation operations to retain the addends in the format carried from the previous operation.
Selected — Configures filter summation operations to convert the input format of the addends to match the summation output format before performing the summation operation. Usually, selecting Cast before sum generates results from the summation that more closely match those found from digital signal processors.
The code generation pane contains options for various implementations of the completed filter design. Depending on your installation, you can generate MATLAB, VHDL, and Verilog code from the designed filter. You can also choose to create or update a Simulink® model from the designed filter. The following section explains these options.
For more information on this option, see Opening the Filter Design HDL Coder UI from the Filter Builder (Filter Design HDL Coder).
Generate MATLAB code based on filter specifications
Generate function that returns your filter as an output
Selecting this option generates a function that designs a
filter object using fdesign
.
Generate function that filters your data
Selecting this option generates a function that takes data as input, and outputs data filtered with the designed filter. The data type of the filter output is set according to the data type settings in the Data Types pane.
Clicking on the Generate MATLAB code button, brings up a Save File dialog. Specify the file name and location, and save. The filter is now contained in an editable file.
Generate Simulink blocks and subsystems from your designed filters
When you click Generate Model, the filter builder generates Simulink blocks and subsystems from your designed filters.
Clicking on the Generate Model button opens the Export to Simulink dialog box.
Block Name — The name for the new subsystem block, set to Filter by default.
Destination —
Current
saves the generated
model to the current Simulink model. New
creates
a new model to contain the generated block.
User Defined
creates a new
model or subsystem at the location specified in
User Defined
.
Overwrite generated 'Filter' block — Overwrites an existing block with the name specified in Block Name. Clear this check box to create a new block with the same name.
Build model using basic elements — Builds the model using only basic blocks.
Optimize for zero gains — Removes all zero-gain blocks from the model.
Optimize for unity gains — Replaces all unity gains with direct connections.
Optimize for negative gains — Removes all negative unity gain blocks, and changes sign at the nearest summation block.
Optimize delay chains — Replaces delay chains made up of n unit delays with a single delay by n.
Optimize for unity scale values — Removes all scale value multiplications by 1 from the filter structure.
Input processing — Specify how the generated filter block or subsystem block processes the input. Depending on the type of filter you are designing, one or both of the following options may be available:
Columns as channels (frame
based)
— The block treats
each column of the input as a separate
channel.
Elements as channels (sample
based)
— The block treats
each element of the input as a separate channel.
For more information about sample-based and frame-based processing, see Sample- and Frame-Based Concepts (DSP System Toolbox).
Realize Model — Builds the model with the set parameters.
When the Use a System object to implement filter check box is selected in the Main pane, the Generate Model button in the Simulink model panel is disabled under the following conditions:
Select Filter response as
Comb
and
Arithmetic on the Data
Types pane as Fixed
point
.
Select Filter response as
Arbitrary Response
,
Impulse response as
IIR
, set Specify
response as to either Magnitudes
and phases
or Frequency
response
, and
Arithmetic on the Data
Types pane as Fixed
point
.
These settings design a dsp.IIRFilter
System object™ with fixed point arithmetic. Generating a Simulink model for fixed point dsp.IIRFilter
object is not supported.
Select your filter response from the filterBuilder
Response Selection main menu.
If you have the DSP System Toolbox software, the following Response Selection menu appears.
Select your desired filter response from the menu and design your filter.
The following sections describe the options available for each response type.
Parameters in this group enable you to specify your filter format, such as the impulse response and the filter order.
This dialog only applies if you have the DSP System Toolbox software. Select either
FIR
or
IIR
from the drop down list,
where FIR
is the default impulse
response. When you choose an impulse response, the design
methods and structures you can use to implement your filter
change accordingly. Arbitrary group delay designs are only
available if Impulse response is
IIR
. Without the DSP System Toolbox, the only available arbitrary response filter
design is FIR.
This dialog only applies if you have the DSP System Toolbox software. Choose
Minimum
or
Specify
. Choosing
Specify
enables the
Order dialog.
This dialog only applies when Order mode
is Specify
. For an FIR design,
specify the filter order. For an IIR design, you can specify an
equal order for the numerator and denominator, or you can
specify different numerator and denominator orders. The default
is equal orders. To specify a different denominator order, check
the Denominator order box. Because the
Signal Processing Toolbox only supports FIR arbitrary-magnitude filters, you
do not have the option to specify a denominator order.
Select the check box and enter the denominator order. This
option is enabled only if IIR
is
selected for Impulse response.
This dialog only applies if you have the DSP System Toolbox software and is only available for FIR filters.
Select Single-rate
,
Decimator
,
Interpolator
, or
Sample-rate converter
. Your
choice determines the type of filter as well as the design
methods and structures that are available to implement your
filter. By default, filterBuilder
specifies
single-rate filters.
Selecting Decimator
or
Interpolator
activates
the Decimation Factor or the
Interpolation Factor options
respectively.
Selecting Sample-rate
converter
activates both
factors.
When you design either a decimator or interpolator, the resulting filter is a bandpass filter that either decimates or interpolates your input signal.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Decimator
or
Sample-rate converter
. The
default factor value is 2 for
Decimator
and 3 for
Sample-rate converter
.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Interpolator
or
Sample-rate converter
. The
default factor value is 2.
Select the number of bands in the filter. Multiband design is available for both FIR and IIR filters.
Specify the response as Amplitudes
,
Magnitudes and phase
,
Frequency response
, or
Group delay
.
Amplitudes
is the only option if you
do not have the DSP System Toolbox software. Group delay
is
only available for IIR designs.
Specify frequency units as either
Normalized
,
Hz
, kHz
,
MHz
, or
GHz
.
Enter the input sampling frequency in the units specified in the Frequency units drop-down box. This option is enabled when Frequency units is set to an option in hertz.
These properties are modified automatically depending on the response chosen in the Specify response as drop-down box. Two or three columns are presented for input. The first column is always Frequencies. The other columns are either Amplitudes, Magnitudes, Phases, or Frequency Response. Enter the corresponding vectors of values for each column.
Frequencies and Amplitudes
— These columns are presented for input if you select
Amplitudes
in the Specify
response as drop-down box.
Frequencies, Magnitudes, and
Phases — These columns are presented for
input if the response chosen in the Specify response
as drop-down box is Magnitudes and
phases
.
Frequencies and Frequency
response — These columns are presented for input
if the response chosen in the Specify response as
drop-down box is Frequency response
.
The options for each design are specific for each design method. In the arbitrary response design, the available options also depend on the Response specifications. This section does not present all of the available options for all designs and design methods.
Select the design method for the filter. Different methods are enabled depending on the defining parameters entered in the previous sections.
Window — Valid when the
Design method is
Frequency Sampling
.
Replace the square brackets with the name of a
window
function or function
handle. For example, 'hamming'
or
@hamming
. If the window function
takes parameters other than the length, use a cell
array. For example, {'kaiser',3.5}
or
{@chebwin,60}
.
Density factor — Valid
when the Design method is
equiripple
. Density
factor controls the density of the frequency grid over
which the design method optimization evaluates your
filter response function. The number of equally spaced
points in the grid is the value you enter for
Density factor times (filter
order + 1).
Increasing the value creates a filter that more closely approximates an ideal equiripple filter but increases the time required to design the filter. The default value of 16 represents a reasonable trade between the accurate approximation to the ideal filter and the time to design the filter.
The default changes to 20 for an IIR arbitrary group delay design.
Phase constraint — Valid
when the Design method is
equiripple
, you have the
DSP System Toolbox installed, and Specify response
as is set to
Amplitudes
. Choose one of
Linear
,
Minimum
, or
Maximum
.
Weights — Uses the weights in Weights to weight the error for a single-band design. If you have multiple frequency bands, the Weights design option changes to B1 Weights, B2 Weights to designate the separate bands. Use Bi Weights to specify weights for the i-th band. The Bi Weights design option is only available when you specify the i-th band as an unconstrained.
Bi forced frequency point
— This option is only available in a multi-band
constrained equiripple design when Specify
response as is
Amplitudes
. Bi
forced frequency point is the frequency
point in the i-th band at which the response is forced
to be zero. The index i
corresponds to the frequency bands in Band
properties. For example, if you specify
two bands in Band properties, you
have B1 forced frequency point and
B2 forced frequency
point.
Norm — Valid only for IIR arbitrary group delay designs. Norm is the norm used in the optimization. The default value is 128, which essentially equals the L-infinity norm. The norm must be even.
Max pole radius — Valid only for IIR arbitrary group delay designs. Constrains the maximum pole radius. The default is 0.999999. Reducing the Max pole radius can produce a transfer function more resistant to quantization.
Init norm — Valid only for IIR arbitrary group delay designs. The initial norm used in the optimization. The default initial norm is 2.
Init numerator — Specifies an initial estimate of the filter numerator coefficients.
Init denominator — Specifies an initial estimate of the filter denominator coefficients. This may be useful in difficult optimization problems. In allpass filters, you only have to specify either the denominator or numerator coefficients. If you specify the denominator coefficients, you can obtain the numerator coefficients.
Select the structure for the filter. The available filter structures depend on the parameters you select for your filter.
This check box appears when you set Filter
type to Single-rate
.
Selecting this check box gives you the choice of using a System object to implement the filter. By default, this check box is
cleared.
This check box no longer appears when you set Filter
type to Interpolator
,
Decimator
, or
Sample-rate converter
. The filter
builder always implements the filter as a System object.
Weighting type — The weighting type
defines the frequency response of the filter. The valid weighting types
are: A, C , C-message, ITU-T 0.41, and ITU-R 468–4 weighting. See
fdesign.audioweighting
(DSP System Toolbox) for
definitions of the weighting types.
Class — Filter class is only applicable for A weighting and C weighting filters. The filter class describes the frequency-dependent tolerances specified in the relevant standards. There are two possible class values: 1 and 2. Class 1 weighting filters have stricter tolerances than class 2 filters. The filter class value does not affect the design. The class value is only used to provide a specification mask in FVTool for the analysis of the filter design.
Impulse response — Impulse response type
as one of IIR
or
FIR
. For A, C , C-message, and ITU-R
468–4 filter, IIR is the only option. For a ITU-T 0.41 weighting filter,
FIR is the only option.
Frequency units — Choose
Hz
, kHz
,
MHz
, or GHz
.
Normalized frequency designs are not supported for audio weighting
filters.
Input sample rate — The sampling frequency
in Frequency units. For example, if
Frequency units is set to
kHz
, setting Input sample
rate to 40 is equivalent to a 40 kHz sampling
frequency.
Design method — Valid design methods
depend on the weighting type. For type A and C weighting filters, the
only valid design type is ANSI S1.42
. This is
an IIR design method that follows ANSI standard S1.42–2001. For a C
message filter, the only valid design method is Bell
41009
, which is an IIR design method following the
Bell System Technical Reference PUB 41009. For a ITU-R 468–4 weighting
filter, you can design an IIR or FIR filter. If you choose an IIR
design, the design method is IIR least
p-norm
. If you choose an FIR design, the design method
choices are: Equiripple
or
Frequency Sampling
. For an ITU-T 0.41
weighting filter, the available FIR design methods are
Equiripple
or Frequency
Sampling
Scale SOS filter coefficients to reduce chance of overflow — Selecting this parameter directs the design to scale the filter coefficients to reduce the chances that the inputs or calculations in the filter overflow and exceed the representable range of the filter. Clearing this option removes the scaling. This parameter applies only to IIR filters.
Structure — For the filter specifications and design method you select, this parameter lists the filter structures available to implement your filter. For audio weighting IIR filter designs, you can choose direct form I or II biquad (SOS). You can also choose to implement these structures in transposed form.
For FIR designs, you can choose direct form, direct-form transposed, direct-form symmetric, direct-form asymmetric structures, or an overlap and add structure.
Use a System object to implement filter — Selecting this check box gives you the choice of using a System object to implement the filter. By default, this check box is cleared. When the current design method or structure is not supported by a System object filter, then this check box is disabled.
Parameters in this group enable you to specify your filter format, such as the impulse response and the filter order.
Select FIR
or
IIR
from the drop-down list,
where FIR
is the default impulse
response. When you choose an impulse response, the design
methods and structures you can use to implement your filter
change accordingly.
Note
The design methods and structures for FIR filters are not the same as the methods and structures for IIR filters.
Select Minimum
(the default) or
Specify
from the drop-down box.
Selecting Specify
enables the
Order option so you can enter the
filter order.
If you have the DSP System Toolbox software installed, you can specify IIR filters with different numerator and denominator orders. The default is equal orders. To specify a different denominator order, check the Denominator order box.
Select Single-rate
,
Decimator
,
Interpolator
, or
Sample-rate converter
. Your
choice determines the type of filter as well as the design
methods and structures that are available to implement your
filter. By default, filterBuilder
specifies
single-rate filters.
Selecting Decimator
or
Interpolator
activates
the Decimation Factor or the
Interpolation Factor options
respectively.
Selecting Sample-rate
converter
activates both
factors.
Enter the filter order. This option is enabled only if you
select Specify
for Order
mode.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Decimator
or
Sample-rate converter
. The
default factor value is 2.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Interpolator
or
Sample-rate converter
. The
default factor value is 2.
The parameters in this group allow you to specify your filter response curve. Graphically, the filter specifications look similar to those shown in the following figure.
In the figure, regions between specification values such as Stopband frequency 1 (Fstop1) and Passband frequency 1 (Fpass1) represent transition regions where the filter response is not explicitly defined.
Select the filter features to use to define the frequency response characteristics. This dialog applies only when Order mode is Specify.
Passband and stopband frequencies
— Define the filter by specifying the frequencies
for the edges for the stop- and passbands.
Passband frequency
— Define
the filter by specifying frequencies for the edges of
the passband.
Stopband frequency
— Define
the filter by specifying frequencies for the edges of
the stopbands.
Half power (3dB) frequency
—
Define the filter response by specifying the locations
of the 3 dB points (IIR filters). The 3-dB point is the
frequency for the point 3 dB below the passband
value.
Half power (3dB) frequencies and passband
width
— Define the filter by
specifying frequencies for the 3-dB points in the filter
response and the width of the passband. (IIR
filters)
Half power (3dB) frequencies and stopband
width
— Define the filter by
specifying frequencies for the 3-dB points in the filter
response and the width of the stopband. (IIR
filters)
Cutoff (6dB) frequency
—
Define the filter response by specifying the locations
of the 6-dB points. The 6-dB point is the frequency for
the point 6 dB below the passband value. (FIR
filters)
Use this parameter to specify whether your frequency settings are
normalized or in absolute frequency. Select Normalized
(0 to 1)
to enter frequencies in normalized form.
This behavior is the default. To enter frequencies in hertz, select
one of the frequency units from the drop-down
list—Hz
,
kHz
, MHz
,
or GHz
. Selecting one of the unit options
enables the Input sample rate parameter.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter input. When you provide an input sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available when you select one of the frequency options from the Frequency units list.
Enter the frequency at the edge of the end of the first stopband. Specify the value in either normalized frequency units or the absolute units you select in Frequency units.
Enter the frequency at the edge of the start of the passband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
Enter the frequency at the edge of the end of the passband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
Enter the frequency at the edge of the start of the second stopband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
The parameters in this group let you specify the filter response in the passbands and stopbands.
Specify as Unconstrained
or
Constrained bands
. You must have the
DSP System Toolbox software to select Constrained
bands
. Selecting Constrained
bands
enables dialogs for both stopbands and the
passband: Stopband attenuation 1,
Stopband attenuation 2, and
Passband ripple. You cannot specify
constraints for all three bands simultaneously.
Setting Magnitude constraints to
Constrained bands
enables the
Wstop and Wpass
options under Design options.
Specify the units for any parameter you provide in magnitude specifications. Select one of the following options from the drop-down list.
Linear
— Specify the
magnitude in linear units.
dB
— Specify the
magnitude in dB (decibels). This is the default
setting.
Squared
— Specify
the magnitude in squared units.
Enter the filter attenuation in the first stopband in the units you choose for Magnitude units, either linear or decibels.
Enter the filter ripple allowed in the passband in the units you choose for Magnitude units, either linear or decibels.
Enter the filter attenuation in the second stopband in the units you choose for Magnitude units, either linear or decibels.
The parameters in this group allow you to specify the design method and
structure that filterBuilder
uses to implement your filter.
Lists the design methods available for the frequency and magnitude specifications you entered. When you change the specifications for a filter, such as changing the impulse response, the methods available to design filters changes as well. The default IIR design method is usually Butterworth, and the default FIR method is equiripple.
Selecting this parameter directs the design to scale the filter coefficients to reduce the chances that the inputs or calculations in the filter overflow and exceed the representable range of the filter. Clearing this option removes the scaling. This parameter applies only to IIR filters.
The options for each design are specific for each design method. This section does not present all of the available options for all designs and design methods. There are many more that you encounter as you select different design methods and filter specifications. The following options represent some of the most common ones available.
Density factor controls the density of the frequency grid over which the design method optimization evaluates your filter response function. The number of equally spaced points in the grid is the value you enter for Density factor times (filter order + 1).
Increasing the value creates a filter that more closely approximates an ideal equiripple filter but increases the time required to design the filter. The default value of 16 represents a reasonable trade between the accurate approximation to the ideal filter and the time to design the filter.
Valid when the Design method is
equiripple
and you have the
DSP System Toolbox installed. Choose one of
Linear
,
Minimum
, or
Maximum
.
This option only applies when you have the DSP System Toolbox software and Order mode is
Minimum
.
Select Any
(default),
Even
, or
Odd
. Selecting
Even
or
Odd
forces the minimum-order
design to be an even or odd order.
Weight for the first stopband.
Passband weight.
Weight for the second stopband.
Valid only for IIR designs. Constrains the maximum pole radius. The default is 1. Reducing the max pole radius can produce a transfer function more resistant to quantization.
Valid only for IIR designs. The initial norm used in the optimization. The default initial norm is 2.
Specifies an initial estimate of the filter numerator coefficients. This may be useful in difficult optimization problems.
Specifies an initial estimate of the filter denominator coefficients. This may be useful in difficult optimization problems.
For the filter specifications and design method you select, this parameter lists the filter structures available to implement your filter. By default, FIR filters use direct-form structure, and IIR filters use direct-form II filters with SOS.
This check box appears when you set Filter
type to Single-rate
.
Selecting this check box gives you the choice of using a System object to implement the filter. By default, this check box is
cleared.
This check box no longer appears when you set Filter
type to Interpolator
,
Decimator
, or
Sample-rate converter
. The filter
builder always implements the filter as a System object.
Parameters in this group enable you to specify your filter format, such as the impulse response and the filter order.
Select FIR
or
IIR
from the drop-down list,
where FIR
is the default impulse
response. When you choose an impulse response, the design
methods and structures you can use to implement your filter
change accordingly.
Note
The design methods and structures for FIR filters are not the same as the methods and structures for IIR filters.
Select Minimum
(the default) or
Specify
from the drop-down list.
Selecting Specify
enables the
Order option so you can enter the
filter order.
If you have the DSP System Toolbox software installed, you can specify IIR filters with different numerator and denominator orders. The default is equal orders. To specify a different denominator order, check the Denominator order box.
Select Single-rate
,
Decimator
,
Interpolator
, or
Sample-rate converter
. Your
choice determines the type of filter as well as the design
methods and structures that are available to implement your
filter. By default, filterBuilder
specifies
single-rate filters.
Selecting Decimator
or
Interpolator
activates
the Decimation Factor or the
Interpolation Factor options
respectively.
Selecting Sample-rate
converter
activates both
factors.
When you design either a decimator or an interpolator, the resulting filter is a bandpass filter that either decimates or interpolates your input signal.
Enter the filter order. This option is enabled only if
Specify
was selected for
Order mode.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Decimator
or
Sample-rate converter
. The
default factor value is 2.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Interpolator
or
Sample-rate converter
. The
default factor value is 2.
The parameters in this group allow you to specify your filter response curve. Graphically, the filter specifications look similar to those shown in the following figure.
Select the filter features to use to define the frequency response characteristics. This dialog applies only when Order mode is Specify.
Passband and stopband
frequencies
— Define the filter
by specifying the frequencies for the edges for the
stop- and passbands.
Passband frequency
—
Define the filter by specifying frequencies for the
edges of the passband.
Stopband frequency
—
Define the filter by specifying frequencies for the
edges of the stopbands.
Half power (3dB) frequency
— Define the filter response by specifying the
locations of the 3 dB points (IIR filters). The 3 dB
point is the frequency for the point 3 dB point below
the passband value.
Half power (3dB) frequencies and passband
width
— Define the filter by
specifying frequencies for the 3 dB points in the filter
response and the width of the passband (IIR filters).
Half power (3dB) frequencies and stopband
width
— Define the filter by
specifying frequencies for the 3 dB points in the filter
response and the width of the stopband (IIR
filters).
Cutoff (6dB) frequency
— Define the filter response by specifying the
locations of the 6-dB points (FIR filters). The 6-dB
point is the frequency for the point 6 dB point below
the passband value.
Use this parameter to specify whether your frequency settings are
normalized or in absolute frequency. Select Normalized
(0 to 1)
to enter frequencies in normalized form.
This behavior is the default. To enter frequencies in absolute
values, select one of the frequency units from the drop-down
list—Hz
,
kHz
, MHz
,
or GHz
. Selecting one of the unit options
enables the Input sample rate parameter.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter input. When you provide an input sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available when you select one of the frequency options from the Frequency units list.
When you design an interpolator, Fs represents the sampling
frequency at the filter output rather than the filter input. This
option is available only when you set Filter
type is
interpolator
.
Enter the frequency at the edge of the end of the first passband. Specify the value in either normalized frequency units or the absolute units you select in Frequency units.
Enter the frequency at the edge of the start of the stopband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
Enter the frequency at the edge of the end of the stopband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
Enter the frequency at the edge of the start of the second passband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
The parameters in this group let you specify the filter response in the passbands and stopbands.
Specify as Unconstrained
or
Constrained bands
. You must have the
DSP System Toolbox software to select Constrained
bands
. Selecting Constrained
bands
enables dialogs for both passbands and the
stopband: Passband ripple 1, Passband
ripple 2, and Stopband
attenuation. You cannot specify constraints for all
three bands simultaneously.
Setting Magnitude constraints to
Constrained bands
enables the
Wstop and Wpass
options under Design options.
Specify the units for any parameter you provide in magnitude specifications. Select one of the following options from the drop-down list.
Linear
— Specify the
magnitude in linear units.
dB
— Specify the
magnitude in decibels (default).
Squared
— Specify
the magnitude in squared units.
Enter the filter ripple allowed in the first passband in the units you choose for Magnitude units, either linear or decibels.
Enter the filter attenuation in the stopband in the units you choose for Magnitude units, either linear or decibels
Enter the filter ripple allowed in the second passband in the units you choose for Magnitude units, either linear or decibels
The parameters in this group allow you to specify the design method and
structure that filterBuilder
uses to implement your filter.
Lists the design methods available for the frequency and magnitude specifications you entered. When you change the specifications for a filter, such as changing the impulse response, the methods available to design filters changes as well. The default IIR design method is usually Butterworth, and the default FIR method is equiripple.
Selecting this parameter directs the design to scale the filter coefficients to reduce the chances that the inputs or calculations in the filter overflow and exceed the representable range of the filter. Clearing this option removes the scaling. This parameter applies only to IIR filters.
The options for each design are specific for each design method. This section does not present all of the available options for all designs and design methods. There are many more that you encounter as you select different design methods and filter specifications. The following options represent some of the most common ones available.
Density factor controls the density of the frequency grid over which the design method optimization evaluates your filter response function. The number of equally spaced points in the grid is the value you enter for Density factor times (filter order + 1).
Increasing the value creates a filter that more closely approximates an ideal equiripple filter but increases the time required to design the filter. The default value of 16 represents a reasonable trade between the accurate approximation to the ideal filter and the time to design the filter.
Valid when the Design method is
equiripple
and you have the
DSP System Toolbox installed. Choose one of
Linear
,
Minimum
, or
Maximum
.
This option only applies when you have the DSP System Toolbox software and Order mode is
Minimum
.
Select Any
(default),
Even
, or
Odd
. Selecting
Even
or
Odd
forces the minimum-order
design to be an even or odd order.
Weight for the first passband.
Stopband weight.
Weight for the second passband.
Specifies that the resulting filter design matches either the
passband or stopband or both bands when you select
passband
or stopband
.
Valid only for IIR designs. Constrains the maximum pole radius. The default is 1. Reducing the max pole radius can produce a transfer function more resistant to quantization.
Valid only for IIR designs. The initial norm used in the optimization. The default initial norm is 2.
Specifies an initial estimate of the filter numerator coefficients. This may be useful in difficult optimization problems.
Specifies an initial estimate of the filter denominator coefficients. This may be useful in difficult optimization problems.
For the filter specifications and design method you select, this parameter lists the filter structures available to implement your filter. By default, FIR filters use direct-form structure, and IIR filters use direct-form II filters with SOS.
This check box appears when you set Filter
type to Single-rate
.
Selecting this check box gives you the choice of using a System object to implement the filter. By default, this check box is
cleared.
This check box no longer appears when you set Filter
type to Interpolator
,
Decimator
, or
Sample-rate converter
. The filter
builder always implements the filter as a System object.
Parameters in this group enable you to specify your CIC filter format, such as the filter type and the differential delay.
Select whether your filter will be a
decimator
or an
interpolator
. Your choice
determines the type of filter and the design methods and
structures that are available to implement your filter.
Selecting decimator
or
interpolator
activates the
Factor option. When you design an
interpolator, you enable the Output sample
rate parameter.
When you design either a decimator or interpolator, the resulting filter is a CIC filter that decimates or interpolates your input signal.
Specify the differential delay of your CIC filter as an integer value greater than or equal to 1. The default value is 1. The differential delay changes the shape, number, and location of nulls in the filter response. Increasing the differential delay increases the sharpness of the nulls and the response between the nulls. In practice, differential delay values of 1 or 2 are the most common.
Specify the decimation or interpolation factor for your filter as an integer value greater than or equal to 1. The default value is 2.
Use this parameter to specify whether your frequency settings are
normalized or in absolute frequency. Select Normalized
(0 to 1)
to enter frequencies in normalized form.
This behavior is the default. To enter frequencies in absolute
values, select one of the frequency units from the drop-down
list—Hz
,
kHz
, MHz
,
or GHz
. Selecting one of the unit options
enables the Input sample rate parameter.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter input. When you provide an input sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available when you select one of the frequency options from the Frequency units list.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter output. When you provide an output sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available only when you design interpolators.
Enter the frequency at the end of the passband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
Specify the units for any parameter you provide in magnitude specifications. Select one of the following options from the drop-down list.
Linear
— Specify the
magnitude in linear units.
dB
— Specify the
magnitude in decibels (default).
Squared
— Specify
the magnitude in squared units.
Enter the filter attenuation in the stopband in the units you choose for Magnitude units, either linear or decibels.
Parameters in this group enable you to specify your filter format, such as the filter order mode and the filter type.
Select Minimum
(the default) or
Specify
from the drop-down list.
Selecting Specify
enables the
Order option (see the following
sections) so you can enter the filter order.
Select Single-rate
,
Decimator
,
Interpolator
, or
Sample-rate converter
. Your
choice determines the type of filter as well as the design
methods and structures that are available to implement your
filter. By default, filterBuilder
specifies
single-rate filters.
Selecting Decimator
or
Interpolator
activates
the Decimation Factor or the
Interpolation Factor options
respectively.
Selecting Sample-rate
converter
activates both
factors.
When you design either a decimator or an interpolator, the resulting filter is a bandpass filter that either decimates or interpolates your input signal.
Enter the filter order. This option is enabled only if
Specify
was selected for
Order mode.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Decimator
or
Sample-rate converter
. The
default factor value is 2.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Interpolator
or
Sample-rate converter
. The
default factor value is 2.
Specify the number of sections in the CIC filter for which you are designing this compensator. Select the number of sections from the drop-down list or enter the number.
Specify the differential delay of your target CIC filter. The
default value is 1
. Most CIC filters use 1 or
2.
The parameters in this group allow you to specify your filter response curve.
Use this parameter to specify whether your frequency settings are
normalized or in absolute frequency. Select Normalized
(0 to 1)
to enter frequencies in normalized form.
This behavior is the default. To enter frequencies in absolute
values, select one of the frequency units from the drop-down
list—Hz
,
kHz
, MHz
,
or GHz
. Selecting one of the unit options
enables the Input sample rate parameter.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter input. When you provide an input sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available when you select one of the frequency options from the Frequency units list.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter output. When you provide an output sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available only when you design interpolators.
Enter the frequency at the end of the passband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
Enter the frequency at the start of the stopband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
The parameters in this group let you specify the filter response in the passbands and stopbands.
Specify the units for any parameter you provide in magnitude specifications. Select one of the following options from the drop-down list.
Linear
— Specify the
magnitude in linear units.
dB
— Specify the
magnitude in decibels (default).
Squared
— Specify
the magnitude in squared units.
Enter the filter ripple allowed in the passband in the units you choose for Magnitude units, either linear or decibels
The parameters in this group allow you to specify the design method and
structure that filterBuilder
uses to implement your filter.
Lists the design methods available for the frequency and magnitude specifications you entered. When you change the specifications for a filter, such as changing the impulse response, the methods available to design filters changes as well. The default IIR design method is usually Butterworth, and the default FIR method is equiripple.
The options for each design are specific for each design method. This section does not present all of the available options for all designs and design methods. There are many more that you encounter as you select different design methods and filter specifications. The following options represent some of the most common ones available.
Density factor controls the density of the frequency grid over which the design method optimization evaluates your filter response function. The number of equally spaced points in the grid is the value you enter for Density factor times (filter order + 1).
Increasing the value creates a filter that more closely approximates an ideal equiripple filter but increases the time required to design the filter. The default value of 16 represents a reasonable trade between the accurate approximation to the ideal filter and the time to design the filter.
To design a filter that is minimum phase, select Minimum phase. Clearing the Minimum phase option removes the phase constraint—the resulting design is not minimum phase.
When you select this parameter, the design method determines
and design the minimum order filter to meet your specifications.
Some filters do not provide this parameter. Select
Any
,
Even
, or
Odd
from the drop-down list to
direct the design to be any minimum order, or minimum even
order, or minimum odd order.
Note
Generally, Minimum order designs are not available for IIR filters.
Specifies that the resulting filter design matches either the
passband or stopband or both bands when you select
passband
or stopband
or both
from the drop-down list.
Stopband shape lets you specify how the stopband changes with increasing frequency. Choose one of the following options:
Flat
— Specifies
that the stopband is flat. The attenuation does not
change as the frequency increases.
Linear
—
Specifies that the stopband attenuation changes
linearly as the frequency increases. Change the
slope of the stopband by setting Stopband
decay.
1/f
— Specifies
that the stopband attenuation changes exponentially
as the frequency increases, where
f
is the frequency. Set the
power (exponent) for the decay in Stopband
decay.
When you set Stopband shape, Stopband decay specifies the amount of decay applied to the stopband. the following conditions apply to Stopband decay based on the value of Stopband Shape:
When you set Stopband shape
to Flat
,
Stopband decay has no affect
on the stopband.
When you set Stopband shape
to Linear
, enter the
slope of the stopband in units of dB/rad/s.
filterBuilder
applies that
slope to the stopband.
When you set Stopband shape
to 1/f
, enter a value for
the exponent n in the relation
(1/f)n
to define the stopband decay.
filterBuilder
applies the
(1/f)n
relation to the stopband to result in an
exponentially decreasing stopband
attenuation.
For the filter specifications and design method you select, this parameter lists the filter structures available to implement your filter. By default, FIR filters use direct-form structure, and IIR filters use direct-form II filters with SOS.
This check box appears when you set Filter
type to Single-rate
.
Selecting this check box gives you the choice of using a System object to implement the filter. By default, this check box is
cleared.
This check box no longer appears when you set Filter
type to Interpolator
,
Decimator
, or
Sample-rate converter
. The filter
builder always implements the filter as a System object.
Parameters in this group enable you to specify the type of comb filter and the number of peaks or notches.
Select Notch
or
Peak
from the drop-down list.
Notch
creates a comb filter that
attenuates a set of harmonically related frequencies.
Peak
creates a comb filter that
amplifies a set of harmonically related frequencies.
Select Order
or Number
of Peaks/Notches
from the drop-down
menu.
Select Order
to enter the desired
filter order in the dialog box. The comb filter
has notches or peaks at increments of 2/
Order
in normalized frequency units.
Select Number of Peaks
or
Number of Notches
to specify the
number of peaks or notches and the Shelving filter
order
.
The Shelving filter order
is a positive
integer that determines the sharpness of the peaks or notches.
Larger values result in sharper peaks or notches.
Parameters in this group enable you to specify the frequency constraints and frequency units.
Select Quality factor
or
Bandwidth
.
Quality factor
is the ratio of the center
frequency of the peak or notch to the bandwidth calculated at the –3
dB point.
Bandwidth
specifies the bandwidth of the peak
or notch. By default the bandwidth is measured at the –3 dB point.
For example, setting the bandwidth equal to 0.1 results in 3 dB
frequencies at normalized frequencies 0.05 above and below the
center frequency of the peak or notch.
Specify the frequency units. The default is normalized frequency. Choosing an option in Hz enables the Input sample rate dialog box.
Specify the units for the magnitude specification and the gain at which the bandwidth is measured. This menu is disabled if you specify a filter order. Select one of the following magnitude units from the drop down list:
dB
— Specify the magnitude in
decibels (default).
Squared
— Specify the magnitude in
squared units.
Bandwidth gain — Specify the gain at which the bandwidth is measured. The default is –3 dB.
The parameters in this group allow you to specify the design method and
structure that filterBuilder
uses to implement your filter.
The IIR Butterworth design is the only option for peaking or notching comb filters.
For the filter specifications and design method you select, this parameter lists the filter structures available to implement your filter.
Selecting this check box gives you the choice of using a System object to implement the filter. By default, the check box is cleared.
Parameters in this group enable you to specify your filter format, such as the impulse response and the filter order. Graphically, the filter specifications look similar to those shown in the following figure.
In the figure, regions between specification values such as Passband frequency (f1) and Stopband frequency (f3) represent transition regions where the filter response is not explicitly defined.
Select Minimum
(the default) or
Specify
from the drop-down list.
Selecting Specify
enables the
Order option (see the following
sections) so you can enter the filter order.
Select Single-rate
,
Decimator
,
Interpolator
, or
Sample-rate converter
. Your
choice determines the type of filter as well as the design
methods and structures that are available to implement your
filter. By default, filterBuilder
specifies
single-rate filters.
Selecting Decimator
or
Interpolator
activates
the Decimation Factor or the
Interpolation Factor options
respectively.
Selecting Sample-rate
converter
activates both
factors.
When you design either a decimator or an interpolator, the resulting filter is a bandpass filter that either decimates or interpolates your input signal.
Enter the filter order. This option is enabled only if
Specify
was selected for
Order mode.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Decimator
or
Sample-rate converter
. The
default factor value is 2.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Interpolator
or
Sample-rate converter
. The
default factor value is 2.
The parameters in this group allow you to specify your filter response curve.
This option is only available when you specify the order of the
filter design. Supported options are
Unconstrained
and
Passband edge and stopband edge
.
Use this parameter to specify whether your frequency settings are
normalized or in absolute frequency. Select Normalized
(0 to 1)
to enter frequencies in normalized form.
This behavior is the default. To enter frequencies in absolute
values, select one of the frequency units from the drop-down
list—Hz
,
kHz
, MHz
,
or GHz
. Selecting one of the unit options
enables the Input sample rate parameter.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter input. When you provide an input sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available when you select one of the frequency options from the Frequency units list.
Enter the frequency at the end of the passband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
Enter the frequency at the start of the stopband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
The parameters in this group let you specify the filter response in the passbands and stopbands.
This option is only available when you specify the order of your
filter design. The options for Magnitude
constraints depend on the value of the
Frequency constraints. If the value of
Frequency constraints is
Unconstrained
, Magnitude
constraints must be
Unconstrained
. If the value of
Frequency constraints is
Passband edge and stopband edge
,
Magnitude constraints can be
Unconstrained
, Passband
ripple
, or Stopband
attenuation
.
Specify the units for any parameter you provide in magnitude specifications. Select one of the following options from the drop-down list.
Linear
— Specify the
magnitude in linear units.
dB
— Specify the
magnitude in decibels (default).
Squared
— Specify
the magnitude in squared units.
Enter the filter ripple allowed in the passband in the units you choose for Magnitude units, either linear or decibels.
Enter the filter attenuation in the second stopband in the units you choose for Magnitude units, either linear or decibels.
The parameters in this group allow you to specify the design method and
structure that filterBuilder
uses to implement your filter.
Lists the design methods available for the frequency and magnitude specifications you entered. When you change the specifications for a filter, such as changing the impulse response, the methods available to design filters changes as well. The default IIR design method is usually Butterworth, and the default FIR method is equiripple.
Selecting this parameter directs the design to scale the filter coefficients to reduce the chances that the inputs or calculations in the filter overflow and exceed the representable range of the filter. Clearing this option removes the scaling. This parameter applies only to IIR filters.
The options for each design are specific for each design method. This section does not present all of the available options for all designs and design methods. There are many more that you encounter as you select different design methods and filter specifications.
Density factor controls the density of the frequency grid over which the design method optimization evaluates your filter response function. The number of equally spaced points in the grid is the value you enter for Density factor times (filter order + 1).
Increasing the value creates a filter that more closely approximates an ideal equiripple filter but increases the time required to design the filter. The default value of 16 represents a reasonable trade between the accurate approximation to the ideal filter and the time to design the filter.
Passband weight. This option is only available for a
specified-order design when Frequency
constraints is equal to Passband
edge and stopband edge
and the
Design method is
Equiripple
.
Stopband weight. This option is only available for a
specified-order design when Frequency
constraints is equal to Passband
edge and stopband edge
and the
Design method is
Equiripple
.
For the filter specifications and design method you select, this parameter lists the filter structures available to implement your filter. By default, FIR filters use direct-form structure, and IIR filters use direct-form II filters with SOS.
This check box appears when you set Filter
type to Single-rate
.
Selecting this check box gives you the choice of using a System object to implement the filter. By default, the check box is
cleared.
This check box no longer appears when you set Filter
type to Interpolator
,
Decimator
, or
Sample-rate converter
. The filter
builder always implements the filter as a System object.
Parameters in this group enable you to specify your filter format, such as the fractional delay and the filter order.
If you choose Specify
for
Order mode, enter your filter order in
this field, or select the order from the drop-down
list.filterBuilder
designs a filter
with the order you specify.
Specify a value between 0 and 1 samples for the filter
fractional delay. The default value is 0.5
samples.
Use this parameter to specify whether your frequency settings are
normalized or in absolute frequency. Select Normalized
(0 to 1)
to enter frequencies in normalized form.
This behavior is the default. To enter frequencies in absolute
values, select one of the frequency units from the drop-down
list—Hz
,
kHz
, MHz
,
or GHz
. Selecting one of the unit options
enables the Input sample rate parameter.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter input. When you provide an input sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available when you select one of the frequency options from the Frequency units list.
Parameters in this group enable you to specify your filter type and order.
Select FIR
or
IIR
from the drop-down list,
where FIR
is the default impulse
response. When you choose an impulse response, the design
methods and structures you can use to implement your filter
change accordingly.
Note
The design methods and structures for FIR filters are not the same as the methods and structures for IIR filters.
Select Minimum
(the default) or
Specify
from the drop-down list.
Selecting Specify
enables the
Order option (see the following
sections) so you can enter the filter order.
Select Single-rate
,
Decimator
, or
Interpolator
. By default,
filterBuilder
specifies single-rate
filters.
When you design either a decimator or an interpolator, the resulting filter is a bandpass filter that decimates or interpolates your input by a factor of two.
Enter the filter order. This option is enabled only if
Specify
was selected for
Order mode.
The parameters in this group allow you to specify your filter response curve. Graphically, the filter specifications for a halfband lowpass filter look similar to those shown in the following figure.
In the figure, the transition region lies between the end of the passband and the start of the stopband. The width is defined explicitly by the value of Transition width.
Use this parameter to specify whether your frequency settings are
normalized or in absolute frequency. Select Normalized
(0 to 1)
to enter frequencies in normalized form.
This behavior is the default. To enter frequencies in absolute
values, select one of the frequency units from the drop-down
list—Hz
,
kHz
, MHz
,
or GHz
. Selecting one of the unit options
enables the Input sample rate parameter.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter input. When you provide an input sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available when you select one of the frequency options from the Frequency units list.
Specify the width of the transition between the end of the passband and the edge of the stopband. Specify the value in normalized frequency units or the absolute units you select in Frequency units.
The parameters in this group let you specify the filter response in the passbands and stopbands.
Specify the units for any parameter you provide in magnitude specifications. Select one of the following options from the drop-down list.
Linear
— Specify the
magnitude in linear units.
dB
— Specify the
magnitude in decibels (default).
Enter the filter attenuation in the stopband in the units you choose for Magnitude units, either linear or decibels.
The parameters in this group allow you to specify the design method and
structure that filterBuilder
uses to implement your filter.
Lists the design methods available for the frequency and
magnitude specifications you entered. For FIR halfband filters,
the available design options are
Equiripple
and
Kaiser window
. For IIR halfband
filters, the available design options are
Butterworth
,
Elliptic
, and IIR
quasi-linear phase
.
The following design options are available for FIR halfband filters when the user specifies an equiripple design:
To design a filter that is minimum phase, select Minimum phase. Clearing the Minimum phase option removes the phase constraint—the resulting design is not minimum phase.
Stopband shape lets you specify how the stopband changes with increasing frequency. Choose one of the following options:
Flat
— Specifies
that the stopband is flat. The attenuation does not
change as the frequency increases.
Linear
—
Specifies that the stopband attenuation changes
linearly as the frequency increases. Change the
slope of the stopband by setting Stopband
decay.
1/f
— Specifies
that the stopband attenuation changes exponentially
as the frequency increases, where
f
is the frequency. Set the
power (exponent) for the decay in Stopband
decay.
When you set Stopband shape, Stopband decay specifies the amount of decay applied to the stopband. the following conditions apply to Stopband decay based on the value of Stopband Shape:
When you set Stopband shape
to Flat
, Stopband
decay has no affect on the
stopband.
When you set Stopband shape
to Linear
, enter the slope of the
stopband in units of dB/rad/s.
filterBuilder
applies that
slope to the stopband.
When you set Stopband shape
to 1/f
, enter a value for the
exponent n in the relation
(1/f)n
to define the stopband decay.
filterBuilder
applies the
(1/f)n
relation to the stopband to result in an
exponentially decreasing stopband
attenuation.
For the filter specifications and design method you select, this parameter lists the filter structures available to implement your filter.
This check box appears when you set Filter
type to Single-rate
.
Selecting this check box gives you the choice of using a System object to implement the filter. By default, the check box is
cleared.
This check box no longer appears when you set Filter
type to either
Interpolator
or
Decimator
. The filter builder always
implements the filter as a System object.
Parameters in this group enable you to specify your filter format, such as the impulse response and the filter order.
Select FIR
or
IIR
from the drop-down list,
where FIR
is the default impulse
response. When you choose an impulse response, the design
methods and structures you can use to implement your filter
change accordingly.
Note
The design methods and structures for FIR filters are not the same as the methods and structures for IIR filters.
Select Minimum
(the default) or
Specify
from the drop-down list.
Selecting Specify
enables the
Order option so you can enter the
filter order.
If your Impulse response is
IIR
, you can specify an equal
order for the numerator and denominator, or different numerator
and denominator orders. The default is equal orders. To specify
a different denominator order, check the Denominator
order box.
This option is only available if you have the DSP System Toolbox software. Select
Single-rate
,
Decimator
,
Interpolator
, or
Sample-rate converter
. Your
choice determines the type of filter as well as the design
methods and structures that are available to implement your
filter. By default, filterBuilder
specifies
single-rate filters.
Selecting Decimator
or
Interpolator
activates
the Decimation Factor or the
Interpolation Factor options
respectively.
Selecting Sample-rate
converter
activates both
factors.
When you design either a decimator or an interpolator, the resulting filter is a highpass filter that either decimates or interpolates your input signal.
Enter the filter order. This option is enabled only if
Specify
was selected for
Order mode.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Decimator
or
Sample-rate converter
. The
default factor value is 2.
Enter the interpolation factor. This option is enabled only if
the Filter type is set to
Interpolator
or
Sample-rate converter
. The
default factor value is 2.
The parameters in this group allow you to specify your filter response curve. Graphically, the filter specifications look similar to those shown in the following figure.
In the figure, the region between specification values Stopband frequency (Fstop) and Passband frequency (Fpass) represents the transition region where the filter response is not explicitly defined.
Select the filter features to use to define the frequency response characteristics. The list contains the following options, when available for the filter specifications.
Stopband edge and passband
edge
— Define the filter by
specifying the frequencies for the edges for the
stopband and passband.
Passband frequency
—
Define the filter by specifying the frequency for the
edge of the passband.
Stopband frequency
—
Define the filter by specifying the frequency for the
edges of the stopband.
Stopband and half power (3dB)
frequencies
— Define the filter
by specifying the stopband edge frequency and the 3-dB
down point (IIR designs).
Half power (3dB) and passband
frequencies
— Define the filter
by specifying the 3-dB down point and passband edge
frequency (IIR designs).
Half power (3dB) frequency
— Define the filter by specifying the frequency
for the 3-dB point (IIR designs or maxflat FIR).
Cutoff (6dB) frequency
— Define the filter by specifying the frequency
for the 6-dB point in the filter response (FIR
designs).
Use this parameter to specify whether your frequency settings are
normalized or in absolute frequency. Select Normalized
(0 to 1)
to enter frequencies in normalized form.
This behavior is the default. To enter frequencies in absolute
values, select one of the frequency units from the drop-down
list—Hz
,
kHz
, MHz
,
or GHz
. Selecting one of the unit options
enables the Input sample rate parameter.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter input. When you provide an input sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available when you select one of the frequency options from the Frequency units list.
Enter the frequency at the of the passband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
Enter the frequency at the start of the stopband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
The parameters in this group let you specify the filter response in the passbands and stopbands.
Specify the units for any parameter you provide in magnitude specifications. Select one of the following options from the drop-down list.
Linear
— Specify the
magnitude in linear units.
dB
— Specify the
magnitude in decibels (default).
Squared
— Specify
the magnitude in squared units.
Enter the filter attenuation in the stopband in the units you choose for Magnitude units, either linear or decibels.
Enter the filter ripple allowed in the passband in the units you choose for Magnitude units, either linear or decibels.
The parameters in this group allow you to specify the design method and
structure that filterBuilder
uses to implement your filter.
Lists the design methods available for the frequency and magnitude specifications you entered. When you change the specifications for a filter, such as changing the impulse response, the methods available to design filters changes as well. The default IIR design method is usually Butterworth, and the default FIR method is equiripple.
Selecting this parameter directs the design to scale the filter coefficients to reduce the chances that the inputs or calculations in the filter overflow and exceed the representable range of the filter. Clearing this option removes the scaling. This parameter applies only to IIR filters.
The options for each design are specific for each design method. This section does not present all of the available options for all designs and design methods. There are many more that you encounter as you select different design methods and filter specifications. The following options represent some of the most common ones available.
Density factor controls the density of the frequency grid over which the design method optimization evaluates your filter response function. The number of equally spaced points in the grid is the value you enter for Density factor times (filter order + 1).
Increasing the value creates a filter that more closely approximates an ideal equiripple filter but increases the time required to design the filter. The default value of 16 represents a reasonable trade between the accurate approximation to the ideal filter and the time to design the filter.
This option only applies when you have the DSP System Toolbox software and when the Design
method is equiripple
.
Select one of Linear
,
Minimum
, or
Maximum
.
Minimum
.Select Any
(default),
Even
, or
Odd
. Selecting
Even
or
Odd
forces the minimum-order
design to be an even or odd order.
Specifies that the resulting filter design matches either the
passband or stopband when you select Passband
or Stopband
.
Stopband shape lets you specify how the stopband changes with increasing frequency. Choose one of the following options:
Flat
— Specifies
that the stopband is flat. The attenuation does not
change as the frequency increases.
Linear
—
Specifies that the stopband attenuation changes
linearly as the frequency increases. Change the
slope of the stopband by setting Stopband
decay.
1/f
— Specifies
that the stopband attenuation changes exponentially
as the frequency increases, where
f
is the frequency. Set the
power (exponent) for the decay in Stopband
decay.
When you set Stopband shape, Stopband decay specifies the amount of decay applied to the stopband. the following conditions apply to Stopband decay based on the value of Stopband Shape:
When you set Stopband shape
to Flat
, Stopband
decay has no affect on the
stopband.
When you set Stopband shape
to Linear
, enter the slope of the
stopband in units of dB/rad/s.
filterBuilder
applies that
slope to the stopband.
When you set Stopband shape
to 1/f
, enter a value for the
exponent n in the relation
(1/f)n
to define the stopband decay.
filterBuilder
applies the
(1/f)n
relation to the stopband to result in an
exponentially decreasing stopband
attenuation.
Passband weight. This option only applies when
Impulse response is
FIR
and Order
mode is
Specify
.
Stopband weight. This option only applies when
Impulse response is
FIR
and Order
mode is
Specify
.
For the filter specifications and design method you select, this parameter lists the filter structures available to implement your filter. By default, FIR filters use direct-form structure, and IIR filters use direct-form II filters with SOS.
This check box appears when you set Filter
type to Single-rate
.
Selecting this check box gives you the choice of using a System object to implement the filter. By default, the check box is
cleared.
This check box no longer appears when you set Filter
type to Interpolator
,
Decimator
, or
Sample-rate converter
. The filter
builder always implements the filter as a System object.
Parameters in this group enable you to specify your filter format, such as the impulse response and the filter order.
Select FIR
or
IIR
from the drop-down list,
where FIR
is the default impulse
response. When you choose an impulse response, the design
methods and structures you can use to implement your filter
change accordingly.
Note
The design methods and structures for FIR filters are not the same as the methods and structures for IIR filters.
This option is only available if you have the DSP System Toolbox software. Select either
Minimum
(the default) or
Specify
from the drop-down list.
Selecting Specify
enables the
Order option (see the following
sections) so you can enter the filter order.
This option is only available if you have the DSP System Toolbox software. Select
Single-rate
,
Decimator
,
Interpolator
, or
Sample-rate converter
. Your
choice determines the type of filter as well as the design
methods and structures that are available to implement your
filter. By default, filterBuilder
specifies
single-rate filters.
Selecting Decimator
or
Interpolator
activates
the Decimation Factor or the
Interpolation Factor options
respectively.
Selecting Sample-rate
converter
activates both
factors.
When you design either a decimator or an interpolator, the resulting filter is a bandpass filter that either decimates or interpolates your input signal.
Enter the filter order. This option is enabled only if
Specify
was selected for
Order mode.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Decimator
or
Sample-rate converter
. The
default factor value is 2.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Interpolator
or
Sample-rate converter
. The
default factor value is 2.
The parameters in this group allow you to specify your filter response curve. Graphically, the filter specifications look similar to those shown in the following figure.
In the figure, the regions between 0 and f1 and between f2 and 1 represent the transition regions where the filter response is explicitly defined by the transition width.
Use this parameter to specify whether your frequency settings are
normalized or in absolute frequency. Select Normalized
(0 to 1)
to enter frequencies in normalized form.
This behavior is the default. To enter frequencies in absolute
values, select one of the frequency units from the drop-down
list—Hz
,
kHz
, MHz
,
or GHz
. Selecting one of the unit options
enables the Input sample rate parameter.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter input. When you provide an input sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available when you select one of the frequency options from the Frequency units list.
Specify the width of the transitions at the ends of the passband. Specify the value in normalized frequency units or the absolute units you select in Frequency units.
The parameters in this group let you specify the filter response in the passbands and stopbands.
Specify the units for any parameter you provide in magnitude specifications. Select one of the following options from the drop-down list.
Linear
— Specify the
magnitude in linear units.
dB
— Specify the
magnitude in decibels (default)
Squared
— Specify
the magnitude in squared units.
Enter the filter ripple allowed in the passband in the units you choose for Magnitude units, either linear or decibels.
The parameters in this group allow you to specify the design method and
structure that filterBuilder
uses to implement your filter.
Lists the design methods available for the frequency and magnitude specifications you entered. When you change the specifications for a filter, such as changing the impulse response, the methods available to design filters changes as well. The default IIR design method is usually Butterworth, and the default FIR method is equiripple.
Selecting this parameter directs the design to scale the filter coefficients to reduce the chances that the inputs or calculations in the filter overflow and exceed the representable range of the filter. Clearing this option removes the scaling. This parameter applies only to IIR filters.
The options for each design are specific for each design method. This section does not present all of the available options for all designs and design methods. There are many more that you encounter as you select different design methods and filter specifications. The following options represent some of the most common ones available.
Density factor controls the density of the frequency grid over which the design method optimization evaluates your filter response function. The number of equally spaced points in the grid is the value you enter for Density factor times (filter order + 1).
Increasing the value creates a filter that more closely approximates an ideal equiripple filter but increases the time required to design the filter. The default value of 16 represents a reasonable trade between the accurate approximation to the ideal filter and the time to design the filter.
This option is only available in a minimum-order design. Specify whether to design a type 3 or a type 4 FIR filter. The filter type is defined as follows:
Type 3 — FIR filter with even order antisymmetric coefficients
Type 4 — FIR filter with odd order antisymmetric coefficients
Select 3
or
4
from the drop-down list.
For the filter specifications and design method you select, this parameter lists the filter structures available to implement your filter. By default, FIR filters use direct-form structure, and IIR filters use direct-form II filters with SOS.
This check box appears when you set Filter
type to Single-rate
.
Selecting this check box gives you the choice of using a System object to implement the filter. By default, the check box is
cleared.
This check box no longer appears when you set Filter
type to Interpolator
,
Decimator
, or
Sample-rate converter
. The filter
builder always implements the filter as a System object.
Parameters in this group enable you to specify your filter format, such as the impulse response and the filter order.
Select Minimum
(the default) or
Specify
from the drop-down list.
Selecting Specify
enables the
Order option (see the following
sections) so you can enter the filter order.
Select Lowpass
or
Highpass
to design an inverse
sinc lowpass or highpass filter.
Select Single-rate
,
Decimator
,
Interpolator
, or
Sample-rate converter
. Your
choice determines the type of filter as well as the design
methods and structures that are available to implement your
filter. By default, filterBuilder
specifies
single-rate filters.
Selecting Decimator
or
Interpolator
activates
the Decimation Factor or the
Interpolation Factor options
respectively.
Selecting Sample-rate
converter
activates both
factors.
When you design either a decimator or an interpolator, the resulting filter is a bandpass filter that either decimates or interpolates your input signal.
Enter the filter order. This option is enabled only if
Specify
was selected for
Order mode.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Decimator
or
Sample-rate converter
. The
default factor value is 2.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Interpolator
or
Sample-rate converter
. The
default factor value is 2.
The parameters in this group allow you to specify your filter response curve. Graphically, the filter specifications look similar to those shown in the following figure.
Regions between specification values such as Passband frequency (Fpass) and Stopband frequency (Fstop) represent transition regions where the filter response is not explicitly defined.
This option is only available when you specify the filter order. The following options are available:
Passband and stopband frequencies
— Define the filter by specifying the frequencies
for the edges for the stop- and passbands.
Passband frequency
— Define
the filter by specifying frequencies for the edges of
the passband.
Stopband frequency
— Define
the filter by specifying frequencies for the edges of
the stopbands.
Cutoff (6dB) frequency
— The
6-dB point is the frequency for the point 6 dB point
below the passband value.
Use this parameter to specify whether your frequency settings are
normalized or in absolute frequency. Select Normalized
(0 to 1)
to enter frequencies in normalized form.
This behavior is the default. To enter frequencies in absolute
values, select one of the frequency units from the drop-down
list—Hz
,
kHz
, MHz
,
or GHz
. Selecting one of the unit options
enables the Input sample rate parameter.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter input. When you provide an input sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available when you select one of the frequency options from the Frequency units list.
Enter the frequency at the end of the passband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
Enter the frequency at the start of the stopband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
The parameters in this group let you specify the filter response in the passbands and stopbands.
Specify the units for any parameter you provide in magnitude specifications. Select one of the following options from the drop-down list.
Linear
— Specify the
magnitude in linear units.
dB
— Specify the
magnitude in decibels (default)
Squared
— Specify
the magnitude in squared units.
Enter the filter ripple allowed in the passband in the units you choose for Magnitude units, either linear or decibels.
Enter the filter attenuation in the stopband in the units you choose for Magnitude units, either linear or decibels.
The parameters in this group allow you to specify the design method and
structure that filterBuilder
uses to implement your filter.
Lists the design methods available for the frequency and magnitude specifications you entered. When you change the specifications for a filter, such as changing the impulse response, the methods available to design filters changes as well. The default IIR design method is usually Butterworth, and the default FIR method is equiripple.
Design Options
The options for each design are specific for each design method. This section does not present all of the available options for all designs and design methods. There are many more that you encounter as you select different design methods and filter specifications. The following options represent some of the most common ones available.
Density factor controls the density of the frequency grid over which the design method optimization evaluates your filter response function. The number of equally spaced points in the grid is the value you enter for Density factor times (filter order + 1).
Increasing the value creates a filter that more closely approximates an ideal equiripple filter but increases the time required to design the filter. The default value of 16 represents a reasonable trade between the accurate approximation to the ideal filter and the time to design the filter.
Available options are Linear
,
Minimum
, and
Maximum
.
Stopband shape lets you specify how the stopband changes with increasing frequency. Choose one of the following options;
Flat
— Specifies
that the stopband is flat. The attenuation does not
change as the frequency increases.
Linear
—
Specifies that the stopband attenuation changes
linearly as the frequency increases. Change the
slope of the stopband by setting Stopband
decay.
1/f
— Specifies
that the stopband attenuation changes exponentially
as the frequency increases, where
f
is the frequency. Set the
power (exponent) for the decay in Stopband
decay.
When you set Stopband shape, Stopband decay specifies the amount of decay applied to the stopband. the following conditions apply to Stopband decay based on the value of Stopband Shape:
When you set Stopband shape
to Flat
, Stopband
decay has no affect on the
stopband.
When you set Stopband shape
to Linear
, enter the slope of the
stopband in units of dB/rad/s.
filterBuilder
applies that
slope to the stopband.
When you set Stopband shape
to 1/f
, enter a value for the
exponent n in the relation
(1/f)n
to define the stopband decay.
filterBuilder
applies the
(1/f)n
relation to the stopband to result in an
exponentially decreasing stopband
attenuation.
A frequency dilation factor. The sinc frequency factor, C , parameterizes the passband magnitude response for a lowpass design through H(ω) = sinc(Cω)^(–P) and for a highpass design through H(ω) = sinc(C(1–ω))^(–P).
Negative power of passband magnitude response. The sinc power, P, parameterizes the passband magnitude response for a lowpass design through H(ω) = sinc(Cω)^(–P) and for a highpass design through H(ω) = sinc(C(1–ω))^(–P).
For the filter specifications and design method you select, this parameter lists the filter structures available to implement your filter. By default, FIR filters use direct-form structure.
This check box appears when you set Filter
type to Single-rate
.
Selecting this check box gives you the choice of using a System object to implement the filter. By default, the check box is
cleared.
This check box no longer appears when you set Filter
type to Interpolator
,
Decimator
, or
Sample-rate converter
. The filter
builder always implements the filter as a System object.
Parameters in this group enable you to specify your filter format, such as the impulse response and the filter order.
Select FIR
or
IIR
from the drop-down list,
where FIR
is the default impulse
response. When you choose an impulse response, the design
methods and structures you can use to implement your filter
change accordingly.
Note
The design methods and structures for FIR filters are not the same as the methods and structures for IIR filters.
Select Minimum
(the default) or
Specify
from the drop-down list.
Selecting Specify
enables the
Order option (see the following
sections) so you can enter the filter order.
If your Impulse response is
IIR
, you can specify an equal
order for the numerator and denominator, or different numerator
and denominator orders. The default is equal orders. To specify
a different denominator order, check the Denominator
order box.
This option is only available if you have the DSP System Toolbox. Select Single-rate
,
Decimator
,
Interpolator
, or
Sample-rate converter
. Your
choice determines the type of filter as well as the design
methods and structures that are available to implement your
filter. By default, filterBuilder
specifies
single-rate filters.
Selecting Decimator
or
Interpolator
activates
the Decimation Factor or the
Interpolation Factor options
respectively.
Selecting Sample-rate
converter
activates both
factors.
When you design either a decimator or an interpolator, the resulting filter is a bandpass filter that either decimates or interpolates your input signal.
Enter the filter order. This option is enabled only if
Specify
was selected for
Order mode.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Decimator
or
Sample-rate converter
. The
default factor value is 2.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Interpolator
or
Sample-rate converter
. The
default factor value is 2.
The parameters in this group allow you to specify your filter response curve. Graphically, the filter specifications look similar to the one shown in the following figure.
In the figure, regions between specification values such as Passband frequency (Fpass) and Stopband frequency (Fstop) represent transition regions where the filter response is not explicitly defined.
Select the filter features to use to define the frequency response characteristics. The list contains the following options, when available for the filter specifications.
Passband and stopband
frequencies
— Define the filter
by specifying the frequencies for the edge of the
stopband and passband.
Passband frequency
—
Define the filter by specifying the frequency for the
edge of the passband.
Stopband frequency
—
Define the filter by specifying the frequency for the
edges of the stopband.
Passband edge and 3dB point
— Define the filter by specifying the passband
edge frequency and the 3-dB down point (IIR
designs).
Half power (3dB) and stopband
frequencies
— Define the filter
by specifying the 3-dB down point and stopband edge
frequency (IIR designs).
Half power (3dB) frequency
— Define the filter by specifying the frequency
for the 3-dB point (IIR designs or maxflat FIR).
Cutoff (6dB) frequency
— Define the filter by specifying the frequency
for the 6-dB point in the filter response (FIR
designs).
Use this parameter to specify whether your frequency settings are
normalized or in absolute frequency. Select Normalized
(0 to 1)
to enter frequencies in normalized form.
This behavior is the default. To enter frequencies in absolute
values, select one of the frequency units from the drop-down
list—Hz
,
kHz
, MHz
,
or GHz
. Selecting one of the unit options
enables the Input sample rate parameter.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter input. When you provide an input sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available when you select one of the frequency options from the Frequency units list.
Enter the frequency at the of the passband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
Enter the frequency at the start of the stopband. Specify the value in either normalized frequency units or the absolute units you select Frequency units.
The parameters in this group let you specify the filter response in the passbands and stopbands.
Specify the units for any parameter you provide in magnitude specifications. Select one of the following options from the drop-down list.
Linear
— Specify the
magnitude in linear units.
dB
— Specify the
magnitude in decibels (default)
Squared
— Specify
the magnitude in squared units.
Enter the filter ripple allowed in the passband in the units you choose for Magnitude units, either linear or decibels.
Enter the filter attenuation in the stopband in the units you choose for Magnitude units, either linear or decibels.
The parameters in this group allow you to specify the design method and
structure that filterBuilder
uses to implement your filter.
Lists the design methods available for the frequency and magnitude specifications you entered. When you change the specifications for a filter, such as changing the impulse response, the methods available to design filters changes as well. The default IIR design method is usually Butterworth, and the default FIR method is equiripple.
Selecting this parameter directs the design to scale the filter coefficients to reduce the chances that the inputs or calculations in the filter overflow and exceed the representable range of the filter. Clearing this option removes the scaling. This parameter applies only to IIR filters.
The options for each design are specific for each design method. This section does not present all of the available options for all designs and design methods. There are many more that you encounter as you select different design methods and filter specifications. The following options represent some of the most common ones available.
Density factor controls the density of the frequency grid over which the design method optimization evaluates your filter response function. The number of equally spaced points in the grid is the value you enter for Density factor times (filter order + 1).
Increasing the value creates a filter that more closely approximates an ideal equiripple filter but increases the time required to design the filter. The default value of 16 represents a reasonable trade between the accurate approximation to the ideal filter and the time to design the filter.
This option only applies when you have the DSP System Toolbox software and when the Design
method is equiripple
.
Select one of Linear
,
Minimum
, or
Maximum
.
Minimum
.Select Any
(default),
Even
, or
Odd
. Selecting
Even
or
Odd
forces the minimum-order
design to be an even or odd order.
Specifies that the resulting filter design matches either the
passband or stopband when you select Passband
or Stopband
.
Stopband shape lets you specify how the stopband changes with increasing frequency. Choose one of the following options:
Flat
— Specifies
that the stopband is flat. The attenuation does not
change as the frequency increases.
Linear
—
Specifies that the stopband attenuation changes
linearly as the frequency increases. Change the
slope of the stopband by setting Stopband
decay.
1/f
— Specifies
that the stopband attenuation changes exponentially
as the frequency increases, where
f
is the frequency. Set the
power (exponent) for the decay in Stopband
decay.
When you set Stopband shape, Stopband decay specifies the amount of decay applied to the stopband. the following conditions apply to Stopband decay based on the value of Stopband Shape:
When you set Stopband shape
to Flat
, Stopband
decay has no affect on the
stopband.
When you set Stopband shape
to Linear
, enter the slope of the
stopband in units of dB/rad/s.
filterBuilder
applies that
slope to the stopband.
When you set Stopband shape
to 1/f
, enter a value for the
exponent n in the relation
(1/f)n
to define the stopband decay.
filterBuilder
applies the
(1/f)n
relation to the stopband to result in an
exponentially decreasing stopband
attenuation.
Passband weight. This option only applies when
Impulse response is
FIR
and Order
mode is
Specify
.
Stopband weight. This option only applies when
Impulse response is
FIR
and Order
mode is
Specify
.
For the filter specifications and design method you select, this parameter lists the filter structures available to implement your filter. By default, FIR filters use direct-form structure, and IIR filters use direct-form II filters with SOS.
This check box appears when you set Filter
type to Single-rate
.
Selecting this check box gives you the choice of using a System object to implement the filter. By default, the check box is
cleared.
This check box no longer appears when you set Filter
type to Interpolator
,
Decimator
, or
Sample-rate converter
. The filter
builder always implements the filter as a System object.
Parameters in this group enable you to specify your filter format, such as the impulse response and the filter order.
Specifies the location of the center of the transition region between the passband and the stopband. The center of the transition region, bw, is calculated using the value for Band:
bw = Fs/(2 × Band).
Select FIR
or
IIR
from the drop-down list,
where FIR
is the default impulse
response. When you choose an impulse response, the design
methods and structures you can use to implement your filter
change accordingly.
Note
The design methods and structures for FIR filters are not the same as the methods and structures for IIR filters.
Select Minimum
(the default) or
Specify
from the drop-down list.
Selecting Specify
enables the
Order option (see the following
sections) so you can enter the filter order.
Select Single-rate
,
Decimator
,
Interpolator
, or
Sample-rate converter
. Your
choice determines the type of filter as well as the design
methods and structures that are available to implement your
filter. By default, filterBuilder
specifies
single-rate filters.
Selecting Decimator
or
Interpolator
activates
the Decimation Factor or the
Interpolation Factor options
respectively.
Selecting Sample-rate
converter
activates both
factors.
When you design either a decimator or an interpolator, the resulting filter is a bandpass filter that either decimates or interpolates your input signal.
Enter the filter order. This option is enabled only if
Specify
was selected for
Order mode.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Decimator
or
Sample-rate converter
. The
default factor value is 2.
Enter the decimation factor. This option is enabled only if
the Filter type is set to
Interpolator
or
Sample-rate converter
. The
default factor value is 2.
The parameters in this group allow you to specify your filter response curve. Graphically, the filter specifications look similar to those shown in the following figure.
In the figure, BW is the width of the transition region and Band determines the location of the center of the region.
Select the filter features to use to define the frequency response characteristics. The list contains the following options, when available for the filter specifications.
Transition width
— Define
the filter using transition width and stopband
attenuation or transition width and order.
Unconstrained
— Define the
filter by specifying the filter order and having no
constraints on the transition width and stopband
attenuation. You can add constraints on the magnitude by
specifying the stopband attenuation.
Use this parameter to specify whether your frequency settings are
normalized or in absolute frequency. Select Normalized
(0 to 1)
to enter frequencies in normalized form.
This behavior is the default. To enter frequencies in absolute
values, select one of the frequency units from the drop-down
list—Hz
,
kHz
, MHz
,
or GHz
. Selecting one of the unit options
enables the Input sample rate parameter.
Fs, specified in the units you selected for Frequency units, defines the sampling frequency at the filter input. When you provide an input sampling frequency, all frequencies in the specifications are in the selected units as well. This parameter is available when you select one of the frequency options from the Frequency units list.
Specify the width of the transition between the end of the passband and the edge of the stopband. Specify the value in normalized frequency units or the absolute units you select in Frequency units.
The parameters in this group let you specify the filter response in the passbands and stopbands.
Specify the units for any parameter you provide in magnitude specifications. Select one of the following options from the drop-down list.
Linear
— Specify the
magnitude in linear units.
dB
— Specify the
magnitude in decibels (default)
Squared
— Specify
the magnitude in squared units.
Enter the filter attenuation in the stopband in the units you choose for Magnitude units, either linear or decibels.
The parameters in this group allow you to specify the design method and
structure that filterBuilder
uses to implement your filter.
Lists the design methods available for the frequency and magnitude specifications you entered. When you change the specifications for a filter, such as changing the impulse response, the methods available to design filters changes as well. The default IIR design method is usually Butterworth, and the default FIR method is equiripple.
Selecting this parameter directs the design to scale the filter coefficients to reduce the chances that the inputs or calculations in the filter overflow and exceed the representable range of the filter. Clearing this option removes the scaling. This parameter applies only to IIR filters.
The options for each design are specific for each design method. This section does not present all of the available options for all designs and design methods. There are many more that you encounter as you select different design methods and filter specifications. The following options represent some of the most common ones available.
Density factor controls the density of the frequency grid over which the design method optimization evaluates your filter response function. The number of equally spaced points in the grid is the value you enter for Density factor times (filter order + 1).
Increasing the value creates a filter that more closely approximates an ideal equiripple filter but increases the time required to design the filter. The default value of 16 represents a reasonable trade between the accurate approximation to the ideal filter and the time to design the filter.
To design a filter that is minimum phase, select Minimum phase. Clearing the Minimum phase option removes the phase constraint—the resulting design is not minimum phase.
When you select this parameter, the design method determines
and designs the minimum order filter to meet your
specifications. Some filters do not provide this parameter.
Select Any
,
Even
, or
Odd
from the drop-down list to
direct the design to be any minimum order, or minimum even
order, or minimum odd order.
Note
Generally, Minimum order designs are not available for IIR filters.
Specifies that the resulting filter design matches either the
passband or stopband or both bands when you select
passband
or stopband
or both
from the drop-down list.
Stopband shape lets you specify how the stopband changes with increasing frequency. Choose one of the following options:
Flat
— Specifies
that the stopband is flat. The attenuation does not
change as the frequency increases.
Linear
—
Specifies that the stopband attenuation changes
linearly as the frequency increases. Change the
slope of the stopband by setting Stopband
decay.
1/f
— Specifies
that the stopband attenuation changes exponentially
as the frequency increases, where
f
is the frequency. Set the
power (exponent) for the decay in Stopband
decay.
When you set Stopband shape, Stopband decay specifies the amount of decay applied to the stopband. the following conditions apply to Stopband decay based on the value of Stopband Shape:
When you set Stopband shape
to Flat
, Stopband
decay has no affect on the
stopband.
When you set Stopband shape
to Linear
, enter the slope of the
stopband in units of dB/rad/s.
filterBuilder
applies that
slope to the stopband.
When you set Stopband shape
to 1/f
, enter a value for the
exponent n in the relation
(1/f)n
to define the stopband decay.
filterBuilder
applies the
(1/f)n
relation to the stopband to result in an
exponentially decreasing stopband
attenuation.
For the filter specifications and design method you select, this parameter lists the filter structures available to implement your filter. By default, FIR filters use direct-form structure, and IIR filters use direct-form II filters with SOS.
This check box appears when you set Filter
type to Single-rate
.
Selecting this check box gives you the choice of using a System object to implement the filter. By default, the check box is
cleared.
This check box no longer appears when you set Filter
type to Interpolator
,
Decimator
, or
Sample-rate converter
. The filter
builder always implements the filter as a System object.
Specify filter order. Possible values are: 4, 6, 8,
10
.
Specify the number of bands per octave. Possible values are:
1, 3, 6, 12, 24
.
Specify frequency units as Hz
or
kHz
.
Specify the input sampling frequency in the frequency units specified previously.
Select from the drop-down list of available center frequency values.
Butterworth is the design method used for this type of filter.
Select the check box to scale the filter coefficients.
Specify filter structure. Choose from:
Direct-form I SOS
Direct-form II SOS
Direct-form I transposed SOS
Direct-form II transposed SOS
Selecting this check box gives you the choice of using a System object to implement the filter. By default, the check box is cleared. When the current design method or structure is not supported by a system object filter, then this check box is disabled.
Select Minimum
to design a minimum
order filter that meets the design specifications, or
Specify
to enter a specific
filter order. The order mode also affects the possible frequency
constraints, which in turn limit the gain specifications. For
example, if you specify a Minimum
order filter, the available frequency constraints are:
Center frequency, bandwidth,
passband width
Center frequency, bandwidth,
stopband width
If you select Specify
, the
available frequency constraints are:
Center frequency,
bandwidth
Center frequency, quality
factor
Shelf type, cutoff frequency, quality
factor
Shelf type, cutoff frequency, shelf
slope parameter
Low frequency, high
frequency
This parameter is enabled only if the Order
mode is set to
Specify
. Enter the filter order in
this text box.
Depending on the filter order, the possible frequency constraints change. Once you choose the frequency constraints, the input boxes in this area change to reflect the selection.
Select the specification to represent the frequency constraints. The following options are available:
Center frequency, bandwidth, passband
width
(available for minimum order
only)
Center frequency, bandwidth, stopband
width
(available for minimum order
only)
Center frequency,
bandwidth
(available for a specified
order only)
Center frequency, quality
factor
(available for a specified
order only)
Shelf type, cutoff frequency, quality
factor
(available for a specified
order only)
Shelf type, cutoff frequency, shelf
slope parameter
(available for a
specified order only)
Low frequency, high
frequency
(available for a specified
order only)
Select the frequency units from the available drop down list
(Normalized, Hz, kHz, MHz, GHz
).
If Normalized
is selected, then the
Input sample rate box is disabled for
input.
Enter the input sampling frequency. This input box is disabled
for input if Normalized
is selected
in the Frequency units input box.
Enter the center frequency in the units specified by the value in Frequency units.
The bandwidth determines the frequency points at which the
filter magnitude is attenuated by the value specified as the
Bandwidth gain in the Gain
specifications section. By default, the
Bandwidth gain defaults to
db(sqrt(.5))
, or –3 dB relative to the
center frequency. The Bandwidth property
only applies when the Frequency constraints
are: Center frequency, bandwidth, passband
width
, Center frequency,
bandwidth, stopband width
, or
Center frequency, bandwidth
.
The passband width determines the frequency points at which
the filter magnitude is attenuated by the value specified as the
Passband gain in the Gain
specifications section. This option is enabled
only if the filter is of minimum order, and the frequency
constraint selected is Center frequency, bandwidth,
passband width
.
The stopband width determines the frequency points at which
the filter magnitude is attenuated by the value specified as the
Stopband gain in the Gain
specifications section. This option is enabled
only if the filter is of minimum order, and the frequency
constraint selected is Center frequency, bandwidth,
stopband width
.
Enter the low frequency cutoff. This option is enabled only if
the filter order is user specified and the frequency constraint
selected is Low frequency, high
frequency
. The filter magnitude is attenuated
by the amount specified in Bandwidth
gain.
Enter the high frequency cutoff. This option is enabled only
if the filter order is user specified and the frequency
constraint selected is Low frequency, high
frequency
. The filter magnitude is attenuated
by the amount specified in Bandwidth
gain.
Depending on the filter order and frequency constraints, the possible gain constraints change. Also, once you choose the gain constraints the input boxes in this area change to reflect the selection.
Select the specification array to represent gain constraints, and remember that not all of these options are available for all configurations. The following is a list of all available options:
Reference, center frequency,
bandwidth, passband
Reference, center frequency,
bandwidth, stopband
Reference, center frequency,
bandwidth, passband, stopband
Reference, center frequency,
bandwidth
Specify the gain units either dB
or
squared
. These units are used for
all gain specifications in the dialog box.
The reference gain determines the level to which the filter magnitude attenuates in Gain units. The reference gain is a floor gain for the filter magnitude response. For example, you may use the reference gain together with the Center frequency gain to leave certain frequencies unattenuated (reference gain of 0 dB) while boosting other frequencies.
Specifies the gain in Gain units at which
the bandwidth is defined. This property applies only when the
Frequency constraints specification
contains a bandwidth
parameter, or is
Low frequency, high
frequency
.
Specify the center frequency in Gain units
The passband gain determines the level in Gain units at which the passband is defined. The passband is determined either by the Passband width value, or the Low frequency and High frequency values in the Frequency specifications section.
The stopband gain is the level in Gain
units at which the stopband is defined. This
property applies only when the Order mode
is minimum and the Frequency constraints
are Center frequency, bandwidth, stopband
width
.
The boost/cut gain applies only when the designing a shelving
filter. Shelving filters include the Shelf
type
parameter in the Frequency
constraints specification. The gain in the
passband of the shelving filter is increased by
Boost/cut gain dB from a
floor gain of 0 dB.
Select the design method from the drop-down list. Different IIR design methods are available depending on the filter constraints you specify.
Select the check box to scale the filter coefficients.
Select filter structure. The possible choices are:
Direct-form I SOS
Direct-form II SOS
Direct-form I transposed SOS
Direct-form II transposed SOS
Selecting this check box gives you the choice of using a System object to implement the filter. By default, the check box is cleared. When the current design method or structure is not supported by a System object filter, then this check box is disabled.
In this area you can specify whether you want to design a peaking filter or a notching filter, as well as the order of the filter.
Select Peak
or
Notch
from the drop-down box.
Enter the filter order. The order must be even.
This group of parameters allows you to specify frequency constraints and units.
Select the frequency constraints for filter specification. There are two choices as follows:
Center frequency and quality
factor
Center frequency and
bandwidth
The frequency units are normalized by default. If you specify
units other than normalized, filterBuilder
assumes that you wish to specify an input sampling frequency,
and enables this input box. The choice of frequency units are:
Normalized (0 to 1), Hz, kHz, MHz,
GHz
.
This input box is enabled if Frequency
units other than Normalized (0 to
1)
are specified. Enter the input sampling
frequency.
Enter the center frequency in the units you specified in Frequency units.
This input box is enabled only when Center
frequency and quality factor
is chosen for the
Frequency Constraints. Enter the
quality factor.
This input box is enabled only when Center
frequency and bandwidth
is chosen for the
Frequency Constraints. Enter the
bandwidth.
This group of parameters allows you to specify the magnitude constraints, as well as their values and units.
Depending on the choice of constraints, the other input boxes are enabled or disabled. Select from four magnitude constraints available:
Unconstrained
Passband ripple
Stopband attenuation
Passband ripple and stopband
attenuation
Select the magnitude units: either dB
or squared
.
This input box is enabled if the magnitude constraints selected
are Passband ripple
or
Passband ripple and stopband
attenuation
. Enter the passband ripple.
This input box is enabled if the magnitude constraints selected
are Stopband attenuation
or
Passband ripple and stopband
attenuation
. Enter the stopband
attenuation.
The parameters in this group allow you to specify the design method and
structure that filterBuilder
uses to implement your filter.
Lists all design methods available for the frequency and magnitude specifications you entered. When you change the specifications for a filter the methods available to design filters changes as well.
Selecting this parameter directs the design to scale the filter coefficients to reduce the chances that the inputs or calculations in the filter overflow and exceed the representable range of the filter. Clearing this option removes the scaling. This parameter applies only to IIR filters.
Lists all available filter structures for the filter specifications and design method you select. The typical options are:
Direct-form I SOS
Direct-form II SOS
Direct-form I transposed SOS
Direct-form II transposed SOS
Selecting this check box gives you the choice of using a System object to implement the filter. By default, the check box is cleared. When the current design method or structure is not supported by a System object filter, then this check box is disabled.
Parameters in this group enable you to specify the shape and length of the filter.
Select the shape of the impulse response from the following options:
Raised Cosine
Square Root Raised
Cosine
Gaussian
This specification is only available for raised cosine and square root raised cosine filters. For these filters, select one of the following options:
Minimum
— This
option will result in the minimum-length filter
satisfying the user-specified Frequency
specifications.
Specify
order
—This option allows the
user to construct a raised cosine or square root
cosine filter of a specified order by entering an
even number in the Order input
box. The length of the impulse response will be
Order+1
.
Specify
symbols
—This option enables the
user to specify the length of the impulse response
in an alternative manner. If Specify
symbols
is chosen, the
Order input box changes to
the Number of symbols input
box.
Specify the oversampling factor. Increasing the oversampling factor guards against aliasing and improves the FIR filter approximation to the ideal frequency response. If Order is specified in Number of symbols, the filter length will be Number of symbols*Samples per symbol+1. The product Number of symbols*Samples per symbol must be an even number.
If a Gaussian filter is specified, the filter length must be specified in Number of symbols and Samples per symbol. The product Number of symbols*Samples per symbol must be an even number. The filter length will be Number of symbols*Samples per symbol+1.
This option is only available if you have the DSP System Toolbox software. Choose Single
rate
, Decimator
,
Interpolator
, or
Sample-rate converter
. If you
select Decimator
or
Interpolator
, the decimation and
interpolation factors default to the value of the
Samples per symbol. If you select
Sample-rate converter
, the
interpolation factor defaults to Samples per
symbol and the decimation factor defaults to
3.
Parameters in this group enable you to specify the frequency response of the filter. For raised cosine and square root raised cosine filters, the frequency specifications include:
The rolloff factor takes values in the range [0,1]. The smaller the rolloff factor, the steeper the transition in the stopband.
The frequency units are normalized by default. If you specify
units other than normalized, filterBuilder
assumes that you wish to specify an input sampling frequency, and
enables this input box. The choice of frequency units are:
Normalized (0 to 1), Hz, kHz, MHz,
GHz
For a Gaussian pulse shape, the available frequency specifications are:
This option allows the user to specify the width of the Gaussian filter. Note that this is independent of the length of the filter. The bandwidth-time product (BT) must be a positive real number. Smaller values of the bandwidth-time product result in larger pulse widths in time and steeper stopband transitions in the frequency response.
The frequency units are normalized by default. If you specify
units other than normalized, filterBuilder
assumes that you wish to specify an input sampling frequency, and
enables this input box. The choice of frequency units are:
Normalized (0 to 1), Hz, kHz, MHz,
GHz
If the Order mode is specified as
Minimum
, the Magnitude units
may be selected from:
dB
— Specify the magnitude in decibels
(default).
Linear
— Specify the magnitude in linear
units.
The only Design method available for FIR pulse-shaping
filters is the Window
method.
For the filter specifications and design method you select, this parameter lists the filter structures available to implement your filter. By default, FIR filters use direct-form structure.
This check box appears when you set Filter
type to Single-rate
.
Selecting this check box gives you the choice of using a System object to implement the filter. By default, the check box is
cleared.
This check box no longer appears when you set Filter
type to Interpolator
,
Decimator
, or
Sample-rate converter
. The filter
builder always implements the filter as a System object.