Intel® Integrated Performance Primitives Developer Guide and Reference

ID 790148
Date 11/07/2023
Public

A newer version of this document is available. Customers should click here to go to the newest version.

Document Table of Contents

IIRGenLowpass, IIRGenHighpass

Computes lowpass and highpass IIR filter coefficients.

Syntax

IppStatus ippsIIRGenLowpass_64f(Ipp64f rFreq, Ipp64f ripple, int order, Ipp64f* pTaps, IppsIIRFilterType filterType, Ipp8u* pBuffer);

IppStatus ippsIIRGenHighpass_64f(Ipp64f rFreq, Ipp64f ripple, int order, Ipp64f* pTaps, IppsIIRFilterType filterType, Ipp8u* pBuffer);

Include Files

ipps.h

Domain Dependencies

Headers: ippcore.h, ippvm.h

Libraries: ippcore.lib, ippvm.lib

Parameters

rFreq

Cutoff frequency, should be in the range (0, 0.5).

ripple

Posible ripple in pass band for ippChebyshev1 type of filter.

order

Order of the filter [1, 12]

pTaps

Pointer to the array where computed tap values are stored.

filterType

Type of the IIR filter, possible values: ippButterworth, ippChebyshev1.

pBuffer

Pointer to the buffer for internal calculations. To get the size of the buffer, use the ippsIIRGenGetBufferSize function.

Description

These functions computes coefficients for lowpass or highpass IIR filters, respectively, with the cutoff frequency rFreq. The parameter filterType specifies the type of the filter. The computed coefficients are stored in the array pTaps. Its length must be at least 2*(order + 1) and the taps arranged in the array as follows:

B0, B1, . . ., Border, A0, A1, . . ., Aorder

Application Notes

Butterworth filters are characterized by a magnitude response that is at most flat in the passband and monotonic overall. Butterworth filters sacrifice rolloff steepness for monotonicity in the passband. Unless the smoothness of the Butterworth filter is needed, Chebyshev1 filter can generally provide steeper rolloff characteristics with a lower filter order. Chebyshev1 type filters are equiripple in the passband and monotonic in the stopband. For ippButterworth filter cutoff frequency is the frequency where the magnitude response of the filter is 2-1/2. For ippChebyshev1 filter cutoff frequency is the frequency at which the magnitude response of the filter is (-ripple) dB. For the functions ippsIIRGenLowpass and ippsIIRGenHighpass, the normalized cutoff frequency rFreq must be a number between 0 and 0.5, where 0.5 corresponds to the Nyquist frequency, π radians per sample. The correspondence between MATLAB's Wn and Intel IPP rFreq is very simple: Wn = 2*rFreq.

Examples:

1) For data sampled at 1000 Hz, create a 9th-order highpass Butterworth filter with cutoff frequency at 300 Hz.

Intel IPP:

Ipp64f pTaps[2*(9+1)];

status = ippsIIRGenHighpass( 300.0/1000.0, 0, 9, pTaps, ippButterworth );

MATLAB:

[b,a] = butter(9,300/500,'high');

2) For data sampled at 1000 Hz, create a 9th-order lowpass Chebyshev1 filter with ripple in the passband of 0.5 dB and a cutoff frequency at 300 Hz.

Intel IPP:

Ipp64f pTaps[2*(9+1)];

status = ippsIIRGenLowpass( 300.0/1000.0, 0.5, 9, pTaps, ippChebyshev1 );

MATLAB:

[b,a] = cheby1(9, 0.5, 300/500);

Return Values

ippStsNoErr

Indicates no error.

ippStsNullPtrErr

Indicates an error when the pTaps pointer is NULL.

ippStsIIRGenOrderErr

Indicates an error when the order is less than 1 or greater than 12.

ippStsFilterFrequencyErr

Indicates an error when the rFreq is out of the range.

See Also