FIR hold interpolator
mfilt.holdinterp
will be removed in a future
release. Use dsp.CICInterpolator
(with NumSections
=
1) instead.
hm = mfilt.holdinterp(l)
hm = mfilt.holdinterp(l)
returns
the object hm
that represents a hold interpolator
with the interpolation factor l
. To work, l
must
be an integer. When you do not include l
in the
calling syntax, it defaults to 2. To perform interpolation by noninteger
amounts, use one of the fractional interpolator objects, such as mfilt.firsrc
.
When you use this hold interpolator, each sample added to the
input signal between existing samples has the value of the most recent
sample from the original signal. Thus you see something like a staircase
profile where the interpolated samples form a plateau between the
previous and next original samples. The example demonstrates this
profile clearly. Compare this to the interpolation process for other
interpolators in the toolbox, such as mfilt.linearinterp
.
Make this filter a fixedpoint or singleprecision filter by
changing the value of the Arithmetic
property
for the filter hm
as follows:
To change to singleprecision filtering, enter
set(hm,'arithmetic','single');
To change to fixedpoint filtering, enter
set(hm,'arithmetic','fixed');
The following table describes the input arguments for creating hm
.
Input Argument  Description 

 Interpolation factor for the filter. 
This section describes the properties for both floatingpoint filters (doubleprecision and singleprecision) and fixedpoint filters.
Every multirate filter object has properties that govern the
way it behaves when you use it. Note that many of the properties are
also input arguments for creating mfilt.holdinterp
objects.
The next table describes each property for an mfilt.interp
filter
object.
Name  Values  Description 


 Specifies the arithmetic the filter uses to process data while filtering. 
 String  Reports the type of filter object. You cannot set this
property — it is always read only and results from your choice
of 
 Integer  Interpolation factor for the filter. 
 '  Determines whether the filter states are restored to zero for each filtering operation. 
 Double or single array  Filter states. 
This table shows the properties associated with the fixedpoint
implementation of the mfilt.holdinterp
filter.
Note The table lists all of the properties that a fixedpoint filter can have. Many of the properties listed are dynamic, meaning they exist only in response to the settings of other properties. To view all of the characteristics for a filter at any time, use info(hm) where 
For further information about the properties of this filter
or any mfilt
object, refer to Multirate Filter Properties.
Name  Values  Description 


 Specifies the arithmetic the filter uses to process data while filtering. 
 String  Reports the type of filter object. You cannot set this
property — it is always read only and results from your choice
of 
 Any positive or negative integer number of bits [15]  Specifies the fraction length the filter uses to interpret input data. 
 Any integer number of bits [16]  Specifies the word length applied to interpret input data. 
 Integer  Interpolation factor for the filter. 

 Determine whether the filter states get restored to zero for each filtering operation 

 Contains the filter states before, during, and after
filter operations. For hold interpolators, the states are always empty
— hold interpolators do not have states. The states use 
Hold interpolators do not have filter coefficients and their filter structure is trivial.
To see the effects of holdbased interpolation, interpolate an input sine wave from 22.05 to 44.1 kHz. Note that each added sample retains the value of the most recent original sample.
l = 2; % Interpolation factor hm = mfilt.holdinterp(l); fs = 22.05e3; % Original sample freq: 22.05 kHz. n = 0:5119; % 5120 samples, 0.232 second long signal x = sin(2*pi*1e3/fs*n); % Original signal, sinusoid at 1 kHz y = filter(hm,x); % 10240 samples, still 0.232 seconds stem(n(1:22)/fs,x(1:22),'filled') % Plot original sampled at % 22.05 kHz hold on % Plot interpolated signal (44.1 kHz) stem(n(1:44)/(fs*l),y(1:44),'r') legend('Original Signal','Interpolated Signal','Location','best'); xlabel('Time (sec)');ylabel('Signal Value')
The following figure shows clearly the step nature of the signal
that comes from interpolating the signal using the hold algorithm
approach. Compare the output to the linear interpolation used in mfilt.linearinterp
.
mfilt.cicinterp
 mfilt.firinterp
 mfilt.firsrc
 mfilt.linearinterp