VE7UWU
/
sdrangel
mirror of https://github.com/f4exb/sdrangel.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

Simplified float halfband filters with unique class for floating point

This commit is contained in:
f4exb 2018-04-29 11:48:46 +02:00
parent 9c49be1313
commit 1213ad2a71
5 changed files with 162 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
sdrbase/CMakeLists.txt
View File

@ -127,8 +127,7 @@ set(sdrbase_HEADERS
dsp/hbfiltertraits.h
dsp/inthalfbandfilter.h
dsp/inthalfbandfilterdb.h
dsp/inthalfbandfilterdbff.h
dsp/inthalfbandfilterdbfi.h
dsp/inthalfbandfilterdbf.h
dsp/inthalfbandfiltereo1.h
dsp/inthalfbandfiltereo1i.h
dsp/inthalfbandfilterst.h

14
sdrbase/dsp/decimatorsff.h
View File

@ -17,7 +17,7 @@
#ifndef SDRBASE_DSP_DECIMATORSFF_H_
#define SDRBASE_DSP_DECIMATORSFF_H_
#include "dsp/inthalfbandfilterdbff.h"
#include "dsp/inthalfbandfilterdbf.h"
#include "export.h"
#define DECIMATORSFF_HB_FILTER_ORDER 64
@ -46,12 +46,12 @@ public:
void decimate64_sup(FSampleVector::iterator* it, const float* buf, qint32 nbIAndQ);
void decimate64_cen(FSampleVector::iterator* it, const float* buf, qint32 nbIAndQ);
IntHalfbandFilterDBFF<double, float, DECIMATORSFF_HB_FILTER_ORDER> m_decimator2; // 1st stages
IntHalfbandFilterDBFF<double, float, DECIMATORSFF_HB_FILTER_ORDER> m_decimator4; // 2nd stages
IntHalfbandFilterDBFF<double, float, DECIMATORSFF_HB_FILTER_ORDER> m_decimator8; // 3rd stages
IntHalfbandFilterDBFF<double, float, DECIMATORSFF_HB_FILTER_ORDER> m_decimator16; // 4th stages
IntHalfbandFilterDBFF<double, float, DECIMATORSFF_HB_FILTER_ORDER> m_decimator32; // 5th stages
IntHalfbandFilterDBFF<double, float, DECIMATORSFF_HB_FILTER_ORDER> m_decimator64; // 6th stages
IntHalfbandFilterDBF<double, float, DECIMATORSFF_HB_FILTER_ORDER> m_decimator2; // 1st stages
IntHalfbandFilterDBF<double, float, DECIMATORSFF_HB_FILTER_ORDER> m_decimator4; // 2nd stages
IntHalfbandFilterDBF<double, float, DECIMATORSFF_HB_FILTER_ORDER> m_decimator8; // 3rd stages
IntHalfbandFilterDBF<double, float, DECIMATORSFF_HB_FILTER_ORDER> m_decimator16; // 4th stages
IntHalfbandFilterDBF<double, float, DECIMATORSFF_HB_FILTER_ORDER> m_decimator32; // 5th stages
IntHalfbandFilterDBF<double, float, DECIMATORSFF_HB_FILTER_ORDER> m_decimator64; // 6th stages
};

2
sdrbase/dsp/decimatorsfi.cpp
View File

@ -14,7 +14,7 @@
// along with this program. If not, see <http://www.gnu.org/licenses/>. //
///////////////////////////////////////////////////////////////////////////////////
#include <dsp/decimatorsfi.h>
#include "decimatorsfi.h"
void DecimatorsFI::decimate1(SampleVector::iterator* it, const float* buf, qint32 nbIAndQ)
{

14
sdrbase/dsp/decimatorsfi.h
View File

@ -17,7 +17,7 @@
#ifndef SDRBASE_DSP_DECIMATORSFI_H_
#define SDRBASE_DSP_DECIMATORSFI_H_
#include "dsp/inthalfbandfilterdbfi.h"
#include "dsp/inthalfbandfilterdbf.h"
#include "export.h"
#define DECIMATORSFI_HB_FILTER_ORDER 64
@ -46,12 +46,12 @@ public:
void decimate64_sup(SampleVector::iterator* it, const float* buf, qint32 nbIAndQ);
void decimate64_cen(SampleVector::iterator* it, const float* buf, qint32 nbIAndQ);
IntHalfbandFilterDBFI<double, float, DECIMATORSFI_HB_FILTER_ORDER> m_decimator2; // 1st stages
IntHalfbandFilterDBFI<double, float, DECIMATORSFI_HB_FILTER_ORDER> m_decimator4; // 2nd stages
IntHalfbandFilterDBFI<double, float, DECIMATORSFI_HB_FILTER_ORDER> m_decimator8; // 3rd stages
IntHalfbandFilterDBFI<double, float, DECIMATORSFI_HB_FILTER_ORDER> m_decimator16; // 4th stages
IntHalfbandFilterDBFI<double, float, DECIMATORSFI_HB_FILTER_ORDER> m_decimator32; // 5th stages
IntHalfbandFilterDBFI<double, float, DECIMATORSFI_HB_FILTER_ORDER> m_decimator64; // 6th stages
IntHalfbandFilterDBF<double, float, DECIMATORSFI_HB_FILTER_ORDER> m_decimator2; // 1st stages
IntHalfbandFilterDBF<double, float, DECIMATORSFI_HB_FILTER_ORDER> m_decimator4; // 2nd stages
IntHalfbandFilterDBF<double, float, DECIMATORSFI_HB_FILTER_ORDER> m_decimator8; // 3rd stages
IntHalfbandFilterDBF<double, float, DECIMATORSFI_HB_FILTER_ORDER> m_decimator16; // 4th stages
IntHalfbandFilterDBF<double, float, DECIMATORSFI_HB_FILTER_ORDER> m_decimator32; // 5th stages
IntHalfbandFilterDBF<double, float, DECIMATORSFI_HB_FILTER_ORDER> m_decimator64; // 6th stages
};

146
sdrbase/dsp/inthalfbandfilterdbf.h Normal file
View File

@ -0,0 +1,146 @@
///////////////////////////////////////////////////////////////////////////////////
// Copyright (C) 2018 F4EXB //
// written by Edouard Griffiths //
// //
// Float half-band FIR based interpolator and decimator //
// This is the double buffer variant //
// //
// This program is free software; you can redistribute it and/or modify //
// it under the terms of the GNU General Public License as published by //
// the Free Software Foundation as version 3 of the License, or //
// //
// This program is distributed in the hope that it will be useful, //
// but WITHOUT ANY WARRANTY; without even the implied warranty of //
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the //
// GNU General Public License V3 for more details. //
// //
// You should have received a copy of the GNU General Public License //
// along with this program. If not, see <http://www.gnu.org/licenses/>. //
///////////////////////////////////////////////////////////////////////////////////
#ifndef INCLUDE_INTHALFBANDFILTER_DBF_H
#define INCLUDE_INTHALFBANDFILTER_DBF_H
#include <stdint.h>
#include "dsp/dsptypes.h"
#include "dsp/hbfiltertraits.h"
#include "export.h"
template<typename AccuType, typename SampleType, uint32_t HBFilterOrder>
class SDRBASE_API IntHalfbandFilterDBF {
public:
IntHalfbandFilterDBF();
void myDecimate(AccuType x1, AccuType y1, AccuType *x2, AccuType *y2)
{
storeSample(x1, y1);
advancePointer();
storeSample(*x2, *y2);
doFIR(x2, y2);
advancePointer();
}
/** Optimized upsampler by 2 not calculating FIR with inserted null samples */
void myInterpolate(qint32 *x1, qint32 *y1, qint32 *x2, qint32 *y2)
{
// insert sample into ring double buffer
m_samplesDB[m_ptr][0] = *x1;
m_samplesDB[m_ptr][1] = *y1;
m_samplesDB[m_ptr + HBFIRFilterTraits<HBFilterOrder>::hbOrder/2][0] = *x1;
m_samplesDB[m_ptr + HBFIRFilterTraits<HBFilterOrder>::hbOrder/2][1] = *y1;
// advance pointer
if (m_ptr < (HBFIRFilterTraits<HBFilterOrder>::hbOrder/2) - 1) {
m_ptr++;
} else {
m_ptr = 0;
}
// first output sample calculated with the middle peak
*x1 = m_samplesDB[m_ptr + (HBFIRFilterTraits<HBFilterOrder>::hbOrder/4) - 1][0];
*y1 = m_samplesDB[m_ptr + (HBFIRFilterTraits<HBFilterOrder>::hbOrder/4) - 1][1];
// second sample calculated with the filter
doInterpolateFIR(x2, y2);
}
protected:
SampleType m_samplesDB[2*(HBFIRFilterTraits<HBFilterOrder>::hbOrder - 1)][2]; // double buffer technique
int m_ptr;
int m_size;
int m_state;
void storeSample(AccuType x, AccuType y)
{
m_samplesDB[m_ptr][0] = x;
m_samplesDB[m_ptr][1] = y;
m_samplesDB[m_ptr + m_size][0] = x;
m_samplesDB[m_ptr + m_size][1] = y;
}
void advancePointer()
{
m_ptr = m_ptr + 1 < m_size ? m_ptr + 1: 0;
}
void doFIR(AccuType *x, AccuType *y)
{
int a = m_ptr + m_size; // tip pointer
int b = m_ptr + 1; // tail pointer
AccuType iAcc = 0;
AccuType qAcc = 0;
for (int i = 0; i < HBFIRFilterTraits<HBFilterOrder>::hbOrder / 4; i++)
{
iAcc += (m_samplesDB[a][0] + m_samplesDB[b][0]) * HBFIRFilterTraits<HBFilterOrder>::hbCoeffsF[i];
qAcc += (m_samplesDB[a][1] + m_samplesDB[b][1]) * HBFIRFilterTraits<HBFilterOrder>::hbCoeffsF[i];
a -= 2;
b += 2;
}
iAcc += m_samplesDB[b-1][0] / 2.0;
qAcc += m_samplesDB[b-1][1] / 2.0;
*x = iAcc; // HB_SHIFT incorrect do not loose the gained bit
*y = qAcc;
}
void doInterpolateFIR(qint32 *x, qint32 *y)
{
qint16 a = m_ptr;
qint16 b = m_ptr + (HBFIRFilterTraits<HBFilterOrder>::hbOrder / 2) - 1;
// go through samples in buffer
AccuType iAcc = 0;
AccuType qAcc = 0;
for (int i = 0; i < HBFIRFilterTraits<HBFilterOrder>::hbOrder / 4; i++)
{
iAcc += (m_samplesDB[a][0] + m_samplesDB[b][0]) * HBFIRFilterTraits<HBFilterOrder>::hbCoeffsF[i];
qAcc += (m_samplesDB[a][1] + m_samplesDB[b][1]) * HBFIRFilterTraits<HBFilterOrder>::hbCoeffsF[i];
a++;
b--;
}
*x = iAcc * SDR_RX_SCALED;
*y = qAcc * SDR_RX_SCALED;
}
};
template<typename AccuType, typename SampleType, uint32_t HBFilterOrder>
IntHalfbandFilterDBF<AccuType, SampleType, HBFilterOrder>::IntHalfbandFilterDBF()
{
m_size = HBFIRFilterTraits<HBFilterOrder>::hbOrder - 1;
for (int i = 0; i < m_size; i++)
{
m_samplesDB[i][0] = 0;
m_samplesDB[i][1] = 0;
}
m_ptr = 0;
m_state = 0;
}
#endif // INCLUDE_INTHALFBANDFILTER_DBF_H