mirror of
https://github.com/f4exb/sdrangel.git
synced 2025-04-11 05:59:32 -04:00
Inf/Sup decimators fix (1): decimators by 2
This commit is contained in:
parent
c3242d618f
commit
0e55accd0f
@ -282,7 +282,8 @@ public:
|
||||
// interleaved I/Q input buffer
|
||||
void decimate1(SampleVector::iterator* it, const T* buf, qint32 len);
|
||||
void decimate2_u(SampleVector::iterator* it, const T* buf, qint32 len);
|
||||
void decimate2_inf(SampleVector::iterator* it, const T* buf, qint32 len);
|
||||
void decimate2_inf_raw(SampleVector::iterator* it, const T* buf, qint32 len);
|
||||
void decimate2_inf(SampleVector::iterator* it, const T* buf, qint32 len);
|
||||
void decimate2_sup(SampleVector::iterator* it, const T* buf, qint32 len);
|
||||
void decimate2_cen(SampleVector::iterator* it, const T* buf, qint32 len);
|
||||
void decimate4_inf(SampleVector::iterator* it, const T* buf, qint32 len);
|
||||
@ -414,90 +415,156 @@ void Decimators<StorageType, T, SdrBits, InputBits>::decimate2_u(SampleVector::i
|
||||
}
|
||||
}
|
||||
|
||||
// No filtering: bad for Rx OK for signal tracking
|
||||
//template<typename StorageType, typename T, uint SdrBits, uint InputBits>
|
||||
//void Decimators<StorageType, T, SdrBits, InputBits>::decimate2_inf(SampleVector::iterator* it, const T* buf, qint32 len)
|
||||
//{
|
||||
// StorageType xreal, yimag;
|
||||
//
|
||||
// for (int pos = 0; pos < len - 7; pos += 8)
|
||||
// {
|
||||
// xreal = (buf[pos+0] - buf[pos+3]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// yimag = (buf[pos+1] + buf[pos+2]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// (**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// (**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// ++(*it);
|
||||
//
|
||||
// xreal = (buf[pos+7] - buf[pos+4]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// yimag = (- buf[pos+5] - buf[pos+6]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// (**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// (**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// ++(*it);
|
||||
// }
|
||||
//}
|
||||
|
||||
//template<typename StorageType, typename T, uint SdrBits, uint InputBits>
|
||||
//void Decimators<StorageType, T, SdrBits, InputBits>::decimate2_inf(SampleVector::iterator* it, const T* bufI, const T* bufQ, qint32 len)
|
||||
//{
|
||||
// StorageType xreal, yimag;
|
||||
//
|
||||
// for (int pos = 0; pos < len - 3; pos += 4)
|
||||
// {
|
||||
// // 0: I[0] 1: Q[0] 2: I[1] 3: Q[1]
|
||||
// xreal = (bufI[pos] - bufQ[pos+1]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// yimag = (bufQ[pos] + bufI[pos+1]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// (**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// (**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// ++(*it);
|
||||
//
|
||||
// // 4: I[2] 5: Q[2] 6: I[3] 7: Q[3]
|
||||
// xreal = (bufQ[pos+3] - bufI[pos+2]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// yimag = (- bufQ[pos+2] - bufI[pos+3]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// (**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// (**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// ++(*it);
|
||||
// }
|
||||
//}
|
||||
|
||||
template<typename StorageType, typename T, uint SdrBits, uint InputBits>
|
||||
void Decimators<StorageType, T, SdrBits, InputBits>::decimate2_inf(SampleVector::iterator* it, const T* buf, qint32 len)
|
||||
{
|
||||
StorageType xreal, yimag;
|
||||
StorageType xreal[2], yimag[2];
|
||||
|
||||
for (int pos = 0; pos < len - 7; pos += 8)
|
||||
{
|
||||
xreal = (buf[pos+0] - buf[pos+3]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
yimag = (buf[pos+1] + buf[pos+2]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
(**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
++(*it);
|
||||
|
||||
xreal = (buf[pos+7] - buf[pos+4]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
yimag = (- buf[pos+5] - buf[pos+6]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
(**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
++(*it);
|
||||
}
|
||||
}
|
||||
|
||||
template<typename StorageType, typename T, uint SdrBits, uint InputBits>
|
||||
void Decimators<StorageType, T, SdrBits, InputBits>::decimate2_inf(SampleVector::iterator* it, const T* bufI, const T* bufQ, qint32 len)
|
||||
{
|
||||
StorageType xreal, yimag;
|
||||
|
||||
for (int pos = 0; pos < len - 3; pos += 4)
|
||||
for (int pos = 0; pos < len - 7; pos += 8)
|
||||
{
|
||||
// 0: I[0] 1: Q[0] 2: I[1] 3: Q[1]
|
||||
xreal = (bufI[pos] - bufQ[pos+1]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
yimag = (bufQ[pos] + bufI[pos+1]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
(**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
xreal[0] = buf[pos+2] << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
yimag[0] = buf[pos+3] << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
xreal[1] = buf[pos+6] << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
yimag[1] = buf[pos+7] << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
|
||||
m_decimator2.myDecimateInf(
|
||||
buf[pos+0] << decimation_shifts<SdrBits, InputBits>::pre2,
|
||||
buf[pos+1] << decimation_shifts<SdrBits, InputBits>::pre2,
|
||||
&xreal[0],
|
||||
&yimag[0],
|
||||
buf[pos+4] << decimation_shifts<SdrBits, InputBits>::pre2,
|
||||
buf[pos+5] << decimation_shifts<SdrBits, InputBits>::pre2,
|
||||
&xreal[1],
|
||||
&yimag[1]);
|
||||
|
||||
(**it).setReal(xreal[0] >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setImag(yimag[0] >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
++(*it);
|
||||
|
||||
// 4: I[2] 5: Q[2] 6: I[3] 7: Q[3]
|
||||
xreal = (bufQ[pos+3] - bufI[pos+2]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
yimag = (- bufQ[pos+2] - bufI[pos+3]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
(**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setReal(xreal[1] >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setImag(yimag[1] >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
++(*it);
|
||||
}
|
||||
}
|
||||
|
||||
// No filtering: bad for Rx OK for signal tracking
|
||||
//template<typename StorageType, typename T, uint SdrBits, uint InputBits>
|
||||
//void Decimators<StorageType, T, SdrBits, InputBits>::decimate2_sup(SampleVector::iterator* it, const T* buf, qint32 len)
|
||||
//{
|
||||
// StorageType xreal, yimag;
|
||||
//
|
||||
// for (int pos = 0; pos < len - 7; pos += 8)
|
||||
// {
|
||||
// xreal = (buf[pos+1] - buf[pos+2]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// yimag = (- buf[pos+0] - buf[pos+3]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// (**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// (**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// ++(*it);
|
||||
//
|
||||
// xreal = (buf[pos+6] - buf[pos+5]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// yimag = (buf[pos+4] + buf[pos+7]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// (**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// (**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// ++(*it);
|
||||
// }
|
||||
//}
|
||||
//
|
||||
//template<typename StorageType, typename T, uint SdrBits, uint InputBits>
|
||||
//void Decimators<StorageType, T, SdrBits, InputBits>::decimate2_sup(SampleVector::iterator* it, const T* bufI, const T* bufQ, qint32 len)
|
||||
//{
|
||||
// StorageType xreal, yimag;
|
||||
//
|
||||
// for (int pos = 0; pos < len - 3; pos += 4)
|
||||
// {
|
||||
// // 0: I[0] 1: Q[0] 2: I[1] 3: Q[1]
|
||||
// xreal = (bufQ[pos] - bufI[pos+1]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// yimag = (- bufI[pos] - bufQ[pos+1]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// (**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// (**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// ++(*it);
|
||||
//
|
||||
// // 4: I[2] 5: Q[2] 6: I[3] 7: Q[3]
|
||||
// xreal = (bufI[pos+3] - bufQ[pos+2]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// yimag = (bufI[pos+2] + bufQ[pos+3]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
// (**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// (**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
// ++(*it);
|
||||
// }
|
||||
//}
|
||||
|
||||
template<typename StorageType, typename T, uint SdrBits, uint InputBits>
|
||||
void Decimators<StorageType, T, SdrBits, InputBits>::decimate2_sup(SampleVector::iterator* it, const T* buf, qint32 len)
|
||||
{
|
||||
StorageType xreal, yimag;
|
||||
StorageType xreal[2], yimag[2];
|
||||
|
||||
for (int pos = 0; pos < len - 7; pos += 8)
|
||||
{
|
||||
xreal = (buf[pos+1] - buf[pos+2]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
yimag = (- buf[pos+0] - buf[pos+3]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
(**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
++(*it);
|
||||
|
||||
xreal = (buf[pos+6] - buf[pos+5]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
yimag = (buf[pos+4] + buf[pos+7]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
(**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
++(*it);
|
||||
}
|
||||
}
|
||||
|
||||
template<typename StorageType, typename T, uint SdrBits, uint InputBits>
|
||||
void Decimators<StorageType, T, SdrBits, InputBits>::decimate2_sup(SampleVector::iterator* it, const T* bufI, const T* bufQ, qint32 len)
|
||||
{
|
||||
StorageType xreal, yimag;
|
||||
|
||||
for (int pos = 0; pos < len - 3; pos += 4)
|
||||
for (int pos = 0; pos < len - 7; pos += 8)
|
||||
{
|
||||
// 0: I[0] 1: Q[0] 2: I[1] 3: Q[1]
|
||||
xreal = (bufQ[pos] - bufI[pos+1]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
yimag = (- bufI[pos] - bufQ[pos+1]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
(**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
xreal[0] = buf[pos+2] << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
yimag[0] = buf[pos+3] << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
xreal[1] = buf[pos+6] << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
yimag[1] = buf[pos+7] << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
|
||||
m_decimator2.myDecimateSup(
|
||||
buf[pos+0] << decimation_shifts<SdrBits, InputBits>::pre2,
|
||||
buf[pos+1] << decimation_shifts<SdrBits, InputBits>::pre2,
|
||||
&xreal[0],
|
||||
&yimag[0],
|
||||
buf[pos+4] << decimation_shifts<SdrBits, InputBits>::pre2,
|
||||
buf[pos+5] << decimation_shifts<SdrBits, InputBits>::pre2,
|
||||
&xreal[1],
|
||||
&yimag[1]);
|
||||
|
||||
(**it).setReal(xreal[0] >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setImag(yimag[0] >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
++(*it);
|
||||
|
||||
// 4: I[2] 5: Q[2] 6: I[3] 7: Q[3]
|
||||
xreal = (bufI[pos+3] - bufQ[pos+2]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
yimag = (bufI[pos+2] + bufQ[pos+3]) << decimation_shifts<SdrBits, InputBits>::pre2;
|
||||
(**it).setReal(xreal >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setImag(yimag >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setReal(xreal[1] >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
(**it).setImag(yimag[1] >> decimation_shifts<SdrBits, InputBits>::post2);
|
||||
++(*it);
|
||||
}
|
||||
}
|
||||
|
@ -572,6 +572,40 @@ public:
|
||||
advancePointer();
|
||||
}
|
||||
|
||||
void myDecimateInf(int32_t x1, int32_t y1, int32_t *x2, int32_t *y2, int32_t x3, int32_t y3, int32_t *x4, int32_t *y4)
|
||||
{
|
||||
storeSample32(-y1, x1);
|
||||
advancePointer();
|
||||
|
||||
storeSample32(-*x2, -*y2);
|
||||
doFIR(x2, y2);
|
||||
advancePointer();
|
||||
|
||||
storeSample32(y3, -x3);
|
||||
advancePointer();
|
||||
|
||||
storeSample32(*x4, *y4);
|
||||
doFIR(x4, y4);
|
||||
advancePointer();
|
||||
}
|
||||
|
||||
void myDecimateSup(int32_t x1, int32_t y1, int32_t *x2, int32_t *y2, int32_t x3, int32_t y3, int32_t *x4, int32_t *y4)
|
||||
{
|
||||
storeSample32(y1, -x1);
|
||||
advancePointer();
|
||||
|
||||
storeSample32(-*x2, -*y2);
|
||||
doFIR(x2, y2);
|
||||
advancePointer();
|
||||
|
||||
storeSample32(-y3, x3);
|
||||
advancePointer();
|
||||
|
||||
storeSample32(*x4, *y4);
|
||||
doFIR(x4, y4);
|
||||
advancePointer();
|
||||
}
|
||||
|
||||
/** Simple zero stuffing and filter */
|
||||
void myInterpolateZeroStuffing(Sample* sample1, Sample* sample2)
|
||||
{
|
||||
|
Loading…
Reference in New Issue
Block a user