mirror of
https://github.com/f4exb/sdrangel.git
synced 2024-11-19 23:01:51 -05:00
214 lines
5.6 KiB
C++
214 lines
5.6 KiB
C++
///////////////////////////////////////////////////////////////////////////////////
|
|
// Copyright (C) 2012 maintech GmbH, Otto-Hahn-Str. 15, 97204 Hoechberg, Germany //
|
|
// written by Christian Daniel //
|
|
// //
|
|
// 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/>. //
|
|
///////////////////////////////////////////////////////////////////////////////////
|
|
|
|
#include <stdio.h>
|
|
#include <errno.h>
|
|
#include "rtlsdrthread.h"
|
|
#include "dsp/samplefifo.h"
|
|
|
|
#define BLOCKSIZE 16384
|
|
|
|
RTLSDRThread::RTLSDRThread(rtlsdr_dev_t* dev, SampleFifo* sampleFifo, QObject* parent) :
|
|
QThread(parent),
|
|
m_running(false),
|
|
m_dev(dev),
|
|
m_convertBuffer(BLOCKSIZE),
|
|
m_sampleFifo(sampleFifo),
|
|
m_samplerate(288000),
|
|
m_log2Decim(4)
|
|
{
|
|
}
|
|
|
|
RTLSDRThread::~RTLSDRThread()
|
|
{
|
|
stopWork();
|
|
}
|
|
|
|
void RTLSDRThread::startWork()
|
|
{
|
|
m_startWaitMutex.lock();
|
|
start();
|
|
while(!m_running)
|
|
m_startWaiter.wait(&m_startWaitMutex, 100);
|
|
m_startWaitMutex.unlock();
|
|
}
|
|
|
|
void RTLSDRThread::stopWork()
|
|
{
|
|
m_running = false;
|
|
wait();
|
|
}
|
|
|
|
void RTLSDRThread::setSamplerate(int samplerate)
|
|
{
|
|
m_samplerate = samplerate;
|
|
}
|
|
|
|
void RTLSDRThread::setLog2Decimation(unsigned int log2_decim)
|
|
{
|
|
m_log2Decim = log2_decim;
|
|
}
|
|
|
|
void RTLSDRThread::run()
|
|
{
|
|
int res;
|
|
|
|
m_running = true;
|
|
m_startWaiter.wakeAll();
|
|
|
|
while(m_running) {
|
|
if((res = rtlsdr_read_async(m_dev, &RTLSDRThread::callbackHelper, this, 32, BLOCKSIZE)) < 0) {
|
|
qCritical("RTLSDRThread: async error: %s", strerror(errno));
|
|
break;
|
|
}
|
|
}
|
|
|
|
m_running = false;
|
|
}
|
|
|
|
void RTLSDRThread::decimate1(SampleVector::iterator* it, const quint8* buf, qint32 len)
|
|
{
|
|
qint16 xreal, yimag;
|
|
for (int pos = 0; pos < len; pos += 2) {
|
|
xreal = buf[pos+0];
|
|
yimag = buf[pos+1];
|
|
Sample s( xreal << 3, yimag << 3 );
|
|
**it = s;
|
|
(*it)++;
|
|
}
|
|
}
|
|
|
|
void RTLSDRThread::decimate2(SampleVector::iterator* it, const quint8* buf, qint32 len)
|
|
{
|
|
qint16 xreal, yimag;
|
|
for (int pos = 0; pos < len - 7; pos += 8) {
|
|
xreal = buf[pos+0] - buf[pos+3];
|
|
yimag = buf[pos+1] + buf[pos+2] - 255;
|
|
Sample s( xreal << 3, yimag << 3 );
|
|
**it = s;
|
|
(*it)++;
|
|
xreal = buf[pos+7] - buf[pos+4];
|
|
yimag = 255 - buf[pos+5] - buf[pos+6];
|
|
Sample t( xreal << 3, yimag << 3 );
|
|
**it = t;
|
|
(*it)++;
|
|
}
|
|
}
|
|
void RTLSDRThread::decimate4(SampleVector::iterator* it, const quint8* buf, qint32 len)
|
|
{
|
|
qint16 xreal, yimag;
|
|
for (int pos = 0; pos < len - 7; pos += 8) {
|
|
xreal = buf[pos+0] - buf[pos+3] + buf[pos+7] - buf[pos+4];
|
|
yimag = buf[pos+1] - buf[pos+5] + buf[pos+2] - buf[pos+6];
|
|
Sample s( xreal << 3, yimag << 3 );
|
|
**it = s;
|
|
(*it)++;
|
|
}
|
|
}
|
|
|
|
void RTLSDRThread::decimate8(SampleVector::iterator* it, const quint8* buf, qint32 len)
|
|
{
|
|
qint16 xreal, yimag;
|
|
for (int pos = 0; pos < len - 15; pos += 8) {
|
|
xreal = buf[pos+0] - buf[pos+3] + buf[pos+7] - buf[pos+4];
|
|
yimag = buf[pos+1] - buf[pos+5] + buf[pos+2] - buf[pos+6];
|
|
Sample s1( xreal << 3, yimag << 3 );
|
|
pos += 8;
|
|
xreal = buf[pos+0] - buf[pos+3] + buf[pos+7] - buf[pos+4];
|
|
yimag = buf[pos+1] - buf[pos+5] + buf[pos+2] - buf[pos+6];
|
|
Sample s2( xreal << 3, yimag << 3 );
|
|
|
|
m_decimator2.myDecimate(&s1, &s2);
|
|
**it = s2;
|
|
(*it)++;
|
|
}
|
|
}
|
|
|
|
void RTLSDRThread::decimate16(SampleVector::iterator* it, const quint8* buf, qint32 len)
|
|
{
|
|
// Offset tuning: 4x downsample and rotate, then
|
|
// downsample 4x more. [ rotate: 0, 1, -3, 2, -4, -5, 7, -6]
|
|
qint16 xreal[4], yimag[4];
|
|
|
|
for (int pos = 0; pos < len - 31; ) {
|
|
for (int i = 0; i < 4; i++) {
|
|
xreal[i] = (buf[pos+0] - buf[pos+3] + buf[pos+7] - buf[pos+4]) << 4;
|
|
yimag[i] = (buf[pos+1] - buf[pos+5] + buf[pos+2] - buf[pos+6]) << 4;
|
|
pos += 8;
|
|
}
|
|
Sample s1( xreal[0], yimag[0] );
|
|
Sample s2( xreal[1], yimag[1] );
|
|
Sample s3( xreal[2], yimag[2] );
|
|
Sample s4( xreal[3], yimag[3] );
|
|
m_decimator2.myDecimate(&s1, &s2);
|
|
m_decimator2.myDecimate(&s3, &s4);
|
|
m_decimator4.myDecimate(&s2, &s4);
|
|
**it = s4;
|
|
(*it)++;
|
|
}
|
|
}
|
|
|
|
|
|
|
|
// Decimate according to specified log2 (ex: log2=4 => decim=16)
|
|
void RTLSDRThread::callback(const quint8* buf, qint32 len)
|
|
{
|
|
SampleVector::iterator it = m_convertBuffer.begin();
|
|
|
|
switch (m_log2Decim)
|
|
{
|
|
case 0:
|
|
decimate1(&it, buf, len);
|
|
break;
|
|
case 1:
|
|
decimate2(&it, buf, len);
|
|
break;
|
|
case 2:
|
|
decimate4(&it, buf, len);
|
|
break;
|
|
case 3:
|
|
decimate8(&it, buf, len);
|
|
break;
|
|
case 4:
|
|
decimate16(&it, buf, len);
|
|
break;
|
|
default:
|
|
break;
|
|
}
|
|
|
|
/*
|
|
if (m_samplerate < 800000)
|
|
decimate4(&it, buf, len);
|
|
else if ((m_samplerate == 1152000)||(m_samplerate == 2048000))
|
|
decimate8(&it, buf, len);
|
|
else
|
|
decimate16(&it, buf, len);
|
|
*/
|
|
|
|
m_sampleFifo->write(m_convertBuffer.begin(), it);
|
|
|
|
if(!m_running)
|
|
rtlsdr_cancel_async(m_dev);
|
|
}
|
|
|
|
void RTLSDRThread::callbackHelper(unsigned char* buf, uint32_t len, void* ctx)
|
|
{
|
|
RTLSDRThread* thread = (RTLSDRThread*)ctx;
|
|
thread->callback(buf, len);
|
|
}
|
|
|