a3x
/
gr-digitalhf
mirror of https://github.com/hb9fxq/gr-digitalhf
1
0
Fork 0
Code Issues Releases Wiki Activity

initial doppler estimate before starting the adaptive DFE

This commit is contained in:
cmayer 2018-10-29 12:25:56 +01:00
parent 56ae39c0ed
commit 4c94787579
4 changed files with 225 additions and 148 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
examples/test_cc.grc
View File

@ -242,7 +242,7 @@
</param>
<param>
<key>label0</key>
<value>PSK</value>
<value>BPSK</value>
</param>
<param>
<key>label1</key>
@ -282,7 +282,7 @@
</param>
<param>
<key>option2</key>
<value>'1'</value>
<value>'2'</value>
</param>
<param>
<key>option3</key>
@ -802,7 +802,7 @@
</param>
<param>
<key>repeat</key>
<value>False</value>
<value>True</value>
</param>
</block>
<block>

264
lib/adaptive_dfe_impl.cc
View File

@ -44,7 +44,17 @@ public:
PyGILState_Release(_state);
}
} ;
boost::python::numpy::ndarray
complex_vector_to_ndarray(std::vector<gr_complex> const& v) {
return boost::python::numpy::from_data
(&v.front(),
boost::python::numpy::dtype::get_builtin<gr_complex>(),
boost::python::make_tuple(v.size()),
boost::python::make_tuple(sizeof(gr_complex)),
boost::python::object());
}
} // anonymous namespace
adaptive_dfe::sptr
adaptive_dfe::make(int sps, // samples per symbol
@ -97,6 +107,7 @@ adaptive_dfe_impl::adaptive_dfe_impl(int sps, // samples per symbol
, _scramble()
, _descrambled_symbols()
, _symbol_counter(0)
, _need_samples(false)
, _df(0)
, _phase(0)
, _b{0.338187046465954, -0.288839024460507}
@ -137,13 +148,12 @@ adaptive_dfe_impl::general_work(int noutput_items,
for (; i<ninput_items[0] && nout < noutput_items; ++i) {
assert(nout < noutput_items);
insert_sample(in[i]);
if (_state == WAIT_FOR_PREAMBLE) {
insert_sample(in[i]);
uint64_t offset = 0;
float phase_est = 0;
if (get_correlation_tag(i, offset, phase_est)) {
_state = DO_FILTER;
_state = INITIAL_DOPPLER_ESTIMATE;
_sample_counter = 0;
_symbol_counter = 0;
// _symbols.clear();
@ -165,113 +175,91 @@ adaptive_dfe_impl::general_work(int noutput_items,
PyErr_Print();
}
}
}
} // WAIT_FOR_PREAMBLE
if (_state == INITIAL_DOPPLER_ESTIMATE) {
// buffer samples and replay them later once the initial doppler estimate is there
if (_samples.size() == _sps * _symbols.size()) {
GILLock gil_lock;
try {
std::vector<gr_complex> const empty_vec;
// initial doppler estimate
if (!update_doppler_information(_physicalLayer.attr("get_doppler")
(complex_vector_to_ndarray(empty_vec),
complex_vector_to_ndarray(_samples)))) {
_state = WAIT_FOR_PREAMBLE;
continue;
}
} catch (boost::python::error_already_set const&) {
PyErr_Print();
}
// (1) correct all samples in the circular buffer with the inital doppler estimate
for (int j=_nB+1; j<_nB+_nF+1; ++j) {
assert(_hist_sample_index+j < 2*(_nB+_nF+1));
_hist_samples[_hist_sample_index+j] *= gr_expj(-_phase);
update_local_oscillator();
}
// (2) insert all buffered samples and run the adaptive filter for them
// instead of pop_front() we first reverse _samples and then insert back() + pop_back()
// O(N) instead of O(N^2)
std::reverse(_samples.begin(), _samples.end());
while (!_samples.empty() && nout < noutput_items) {
insert_sample(_samples.back());
_sample_counter += 1;
_samples.pop_back();
if ((_sample_counter%_sps) == 0)
out[nout++] = filter();
}
if (_samples.empty()) {
_state = DO_FILTER;
} else {
_state = INITIAL_DOPPLER_ESTIMATE_CONTINUE;
}
continue;
}
_samples.push_back(in[i]);
} // INITIAL_DOPPLER_ESTIMATE_CONTINUE
if (_state == INITIAL_DOPPLER_ESTIMATE_CONTINUE) {
std::cout << "INITIAL_DOPPLER_ESTIMATE_CONTINUE\n";
while (!_samples.empty() && nout < noutput_items) {
insert_sample(_samples.back());
_sample_counter += 1;
_samples.pop_back();
if ((_sample_counter%_sps) == 0)
out[nout++] = filter();
}
if (_samples.empty()) {
_state = DO_FILTER;
} else {
_state = INITIAL_DOPPLER_ESTIMATE_CONTINUE;
}
continue;
} // INITIAL_DOPPLER_ESTIMATE_CONTINUE
if (_state == DO_FILTER) {
gr_complex dot_samples = 0;
// volk_32fc_x2_dot_prod_32fc(&dot_samples,
// _hist_samples+_hist_sample_index,
// _taps_samples,
// _nB+_nF+1);
// if (_sample_counter < 80*5)
// std::cout << "SAMPLE " << _sample_counter << " " << dot_samples << std::endl;
gr_complex filter_output = dot_samples;
_samples.push_back(_hist_samples[_hist_sample_index+_nB+1]);
if ((_sample_counter%_sps) == 0) {
if (_symbol_counter == _symbols.size()) {
_symbol_counter = 0;
GILLock gil_lock;
try {
boost::python::numpy::ndarray sy =
boost::python::numpy::from_data(&_descrambled_symbols.front(),
boost::python::numpy::dtype::get_builtin<gr_complex>(),
boost::python::make_tuple(_descrambled_symbols.size()),
boost::python::make_tuple(sizeof(gr_complex)),
boost::python::object());
boost::python::numpy::ndarray sa =
boost::python::numpy::from_data(&_samples.front(),
boost::python::numpy::dtype::get_builtin<gr_complex>(),
boost::python::make_tuple(_samples.size()),
boost::python::make_tuple(sizeof(gr_complex)),
boost::python::object());
update_doppler_information(_physicalLayer.attr("get_doppler")
(complex_vector_to_ndarray(_descrambled_symbols),
complex_vector_to_ndarray(_samples)));
_samples.clear();
update_doppler_information(_physicalLayer.attr("get_doppler")(sy, sa));
update_frame_information(_physicalLayer.attr("get_frame")());
} catch (boost::python::error_already_set const&) {
PyErr_Print();
}
}
gr_complex known_symbol = _symbols[_symbol_counter];
bool is_known = true;
filter_output = 0;
#if 1
volk_32fc_x2_dot_prod_32fc(&filter_output,
_hist_samples+_hist_sample_index,
_taps_samples,
_nB+_nF+1);
#else
for (int l=0; l<_nB+_nF+1; ++l) {
assert(_hist_sample_index+l < 2*(_nB+_nF+1));
filter_output += _hist_samples[_hist_sample_index+l]*_taps_samples[l];
}
#endif
gr_complex dot_symbols=0;
for (int l=0; l<_nW; ++l) {
assert(_hist_symbol_index+l < 2*_nW);
dot_symbols += _hist_symbols[_hist_symbol_index+l]*_taps_symbols[l];
}
filter_output += dot_symbols;
if (std::abs(known_symbol) < 1e-5) { // not known
is_known = false;
gr_complex descrambled_filter_output = std::conj(_scramble[_symbol_counter]) * filter_output;
gr::digital::constellation_sptr constell = _constellations[_constellation_index];
unsigned int jc = constell->decision_maker(&descrambled_filter_output);
gr_complex descrambled_symbol = 0;
constell->map_to_points(jc, &descrambled_symbol);
// make soft decisions
float const err = std::abs(descrambled_filter_output - descrambled_symbol);
_npwr_counter[_constellation_index] += (_npwr_counter[_constellation_index] < _npwr_max_time_constant);
float const alpha = 1.0f/_npwr_counter[_constellation_index];
_npwr[_constellation_index] = (1-alpha)*_npwr[_constellation_index] + alpha*err;
std::vector<float> soft_dec = constell->calc_soft_dec(descrambled_filter_output, _npwr[_constellation_index]);
// std::cout << "soft_dec " << _npwr[_constellation_index] << " : ";
// for (int k=0; k<soft_dec.size(); ++k) {
// std::cout << soft_dec[k] << " ";
// }
// std::cout << "\n";
known_symbol = _scramble[_symbol_counter] * descrambled_symbol;
}
gr_complex err = filter_output - known_symbol;
int jMax=0;
float tMax=0;
for (int j=0; j<_nB+_nF+1; ++j) {
assert(_hist_sample_index+j < 2*(_nB+_nF+1));
_taps_samples[j] -= _mu*err*std::conj(_hist_samples[_hist_sample_index+j]);
// if (std::abs(_taps_samples[j]) > tMax) {
// tMax = std::abs(_taps_samples[j]);
// jMax = j;
// }
}
// std::cout << "taps_max: " << jMax << " " << tMax << std::endl;
for (int j=0; j<_nW; ++j) {
assert(_hist_symbol_index+j < 2*_nW);
_taps_symbols[j] -= _mu*err*std::conj(_hist_symbols[_hist_symbol_index+j]) + _alpha*_taps_symbols[j];
}
// if (_sample_counter < 80*5)
// std::cout << "filter: " << _symbol_counter << " " << _sample_counter << " " << filter_output << " " << known_symbol << " " << std::abs(err) << std::endl;
if (is_known || true) {
_hist_symbols[_hist_symbol_index] = _hist_symbols[_hist_symbol_index + _nW] = known_symbol;
if (++_hist_symbol_index == _nW)
_hist_symbol_index = 0;
}
_descrambled_symbols[_symbol_counter] = filter_output*std::conj(_scramble[_symbol_counter]);
out[nout++] = filter_output*std::conj(_scramble[_symbol_counter]);
++_symbol_counter;
out[nout++] = filter();
}
insert_sample(in[i]);
if (_need_samples)
_samples.push_back(_hist_samples[_hist_sample_index+_nB+1]);
_sample_counter += 1;
}
}
} // DO_FILTER
} // next input sample
consume(0, i);
@ -311,7 +299,7 @@ bool adaptive_dfe_impl::start()
try {
boost::python::object module = boost::python::import(boost::python::str("digitalhf.physical_layer." + _py_module_name));
boost::python::object PhysicalLayer = module.attr("PhysicalLayer");
_physicalLayer = PhysicalLayer();
_physicalLayer = PhysicalLayer(_sps);
update_constellations(_physicalLayer.attr("get_constellations")());
} catch (boost::python::error_already_set const&) {
PyErr_Print();
@ -331,10 +319,71 @@ bool adaptive_dfe_impl::stop()
VOLK_SAFE_DELETE(_hist_symbols);
return true;
}
gr_complex adaptive_dfe_impl::filter() {
gr_complex filter_output = 0;
volk_32fc_x2_dot_prod_32fc(&filter_output,
_hist_samples+_hist_sample_index,
_taps_samples,
_nB+_nF+1);
gr_complex dot_symbols=0;
for (int l=0; l<_nW; ++l) {
assert(_hist_symbol_index+l < 2*_nW);
dot_symbols += _hist_symbols[_hist_symbol_index+l]*_taps_symbols[l];
}
filter_output += dot_symbols;
gr_complex known_symbol = _symbols[_symbol_counter];
bool const is_known = std::abs(known_symbol) > 1e-5;
if (not is_known) { // not known
gr_complex const descrambled_filter_output = std::conj(_scramble[_symbol_counter]) * filter_output;
gr::digital::constellation_sptr constell = _constellations[_constellation_index];
unsigned int jc = constell->decision_maker(&descrambled_filter_output);
gr_complex descrambled_symbol = 0;
constell->map_to_points(jc, &descrambled_symbol);
// make soft decisions
float const err = std::abs(descrambled_filter_output - descrambled_symbol);
_npwr_counter[_constellation_index] += (_npwr_counter[_constellation_index] < _npwr_max_time_constant);
float const alpha = 1.0f/_npwr_counter[_constellation_index];
_npwr[_constellation_index] = (1-alpha)*_npwr[_constellation_index] + alpha*err;
std::vector<float> soft_dec = constell->calc_soft_dec(descrambled_filter_output, _npwr[_constellation_index]);
// std::cout << "soft_dec " << _npwr[_constellation_index] << " : ";
// for (int k=0; k<soft_dec.size(); ++k) {
// std::cout << soft_dec[k] << " ";
// }
// std::cout << "\n";
known_symbol = _scramble[_symbol_counter] * descrambled_symbol;
}
gr_complex err = filter_output - known_symbol;
int jMax=0;
float tMax=0;
for (int j=0; j<_nB+_nF+1; ++j) {
assert(_hist_sample_index+j < 2*(_nB+_nF+1));
_taps_samples[j] -= _mu*err*std::conj(_hist_samples[_hist_sample_index+j]);
// if (std::abs(_taps_samples[j]) > tMax) {
// tMax = std::abs(_taps_samples[j]);
// jMax = j;
// }
}
// std::cout << "taps_max: " << jMax << " " << tMax << std::endl;
for (int j=0; j<_nW; ++j) {
assert(_hist_symbol_index+j < 2*_nW);
_taps_symbols[j] -= _mu*err*std::conj(_hist_symbols[_hist_symbol_index+j]) + _alpha*_taps_symbols[j];
}
// if (_sample_counter < 80*5)
// std::cout << "filter: " << _symbol_counter << " " << _sample_counter << " " << filter_output << " " << known_symbol << " " << std::abs(err) << std::endl;
if (is_known || true) {
_hist_symbols[_hist_symbol_index] = _hist_symbols[_hist_symbol_index + _nW] = known_symbol;
if (++_hist_symbol_index == _nW)
_hist_symbol_index = 0;
}
_descrambled_symbols[_symbol_counter] = filter_output*std::conj(_scramble[_symbol_counter]);
return filter_output*std::conj(_scramble[_symbol_counter++]);
}
void adaptive_dfe_impl::set_mode(std::string mode) {
gr::thread::scoped_lock lock(d_setlock);
std::cout << "adaptive_dfe_impl::stop()" << std::endl;
std::cout << "adaptive_dfe_impl::set_mode " << mode << std::endl;
GILLock gil_lock;
try {
_physicalLayer.attr("set_mode")(mode);
@ -342,7 +391,6 @@ void adaptive_dfe_impl::set_mode(std::string mode) {
PyErr_Print();
return;
}
update_constellations(_physicalLayer.attr("get_constellations")());
}
void adaptive_dfe_impl::update_constellations(boost::python::object obj)
@ -368,10 +416,10 @@ void adaptive_dfe_impl::update_constellations(boost::python::object obj)
_npwr_counter[i] = 0;
}
}
void adaptive_dfe_impl::update_frame_information(boost::python::object obj)
bool adaptive_dfe_impl::update_frame_information(boost::python::object obj)
{
int const n = boost::python::extract<int>(obj.attr("__len__")());
assert(n==2);
assert(n==3);
boost::python::numpy::ndarray array = boost::python::numpy::array(obj[0]);
char const* data = array.get_data();
int const m = array.shape(0);
@ -384,9 +432,11 @@ void adaptive_dfe_impl::update_frame_information(boost::python::object obj)
std::memcpy(&_scramble[i], data+16*i+8, sizeof(gr_complex));
// std::cout << "get_frame " << i << " " << _symbols[i] << " " << _scramble[i] << std::endl;
}
_constellation_index = boost::python::extract<int>(obj[1]);
_constellation_index = boost::python::extract<int> (obj[1]);
_need_samples = boost::python::extract<bool>(obj[2]);
return true;
}
void adaptive_dfe_impl::update_doppler_information(boost::python::object obj)
bool adaptive_dfe_impl::update_doppler_information(boost::python::object obj)
{
int const n = boost::python::extract<int>(obj.attr("__len__")());
assert(n==2);
@ -398,10 +448,11 @@ void adaptive_dfe_impl::update_doppler_information(boost::python::object obj)
std::fill_n(_hist_samples, 2*(_nB+_nF+1), gr_complex(0));
_hist_sample_index = 0;
_sample_counter = 0;
return;
return false;
}
float const doppler = boost::python::extract<float>(obj[1]);
update_pll(doppler);
return true;
}
void adaptive_dfe_impl::update_pll(float doppler) {
@ -422,8 +473,9 @@ void adaptive_dfe_impl::insert_sample(gr_complex z) {
_hist_samples[_hist_sample_index] = _hist_samples[_hist_sample_index+_nB+_nF+1] = z * gr_expj(-_phase);
if (++_hist_sample_index == _nB+_nF+1)
_hist_sample_index = 0;
// local oscillator update
update_local_oscillator();
}
void adaptive_dfe_impl:: update_local_oscillator() {
_phase += _df;
if (_phase > M_PI)
_phase -= 2*M_PI;

11
lib/adaptive_dfe_impl.h
View File

@ -63,6 +63,8 @@ private:
std::vector<gr_complex> _descrambled_symbols;
int _symbol_counter;
bool _need_samples;
// PLL for doppler tracking
float _df; // frequency offset in radians per sample
float _phase; // accumulated phase for frequency correction
@ -71,12 +73,17 @@ private:
enum state {
WAIT_FOR_PREAMBLE,
INITIAL_DOPPLER_ESTIMATE,
INITIAL_DOPPLER_ESTIMATE_CONTINUE,
DO_FILTER
} _state;
void update_constellations(boost::python::object obj);
void update_frame_information(boost::python::object obj);
void update_doppler_information(boost::python::object obj);
bool update_frame_information(boost::python::object obj);
bool update_doppler_information(boost::python::object obj);
void update_local_oscillator();
gr_complex filter();
void insert_sample(gr_complex z);
void update_pll(float doppler);

92
python/physical_layer/STANAG_4285.py
View File

@ -6,56 +6,72 @@ from gnuradio import digital
class PhysicalLayer(object):
"""Physical layer description for STANAG 4285"""
def __init__(self, mode=0):
"""For STANAG 4258 the mode has to be set manually: mode=0 -> BPSK, mode=1 -> QPSK, mode=2 -> 8PSK"""
self._constellations = [PhysicalLayer.make_psk(2, [0,1]),
PhysicalLayer.make_psk(4, [0,1,3,2]),
PhysicalLayer.make_psk(8, [1,0,2,3,6,7,5,4])]
self._preamble = [PhysicalLayer.get_preamble(), 0] ## BPSK
self._data = [PhysicalLayer.get_data(), mode] ## according to the mode
self._counter = 0
self._is_first_frame = True
MODE_BPSK=0
MODE_QPSK=1
MODE_8PSK=2
def __init__(self, sps):
"""intialization"""
self._sps = sps
self._mode = self.MODE_BPSK
self._frame_counter = 0
self._is_first_frame = True
self._constellations = [self.make_psk(2, [0,1]),
self.make_psk(4, [0,1,3,2]),
self.make_psk(8, [1,0,2,3,6,7,5,4])]
self._preamble = self.get_preamble()
self._data = self.get_data()
def set_mode(self, mode):
"""For STANAG 4258 the mode has to be set manually: mode=0 -> BPSK, mode=1 -> QPSK, mode=2 -> 8PSK"""
"""set phase modultation type"""
print('set_mode', mode)
self._data[1] = int(mode)
self._mode = int(mode)
def get_constellations(self):
return self._constellations
def get_frame(self):
"""returns the known+unknown symbols and scrambling"""
print('-------------------- get_frame --------------------',self._counter)
return self._preamble if self._counter == 0 else self._data
"""returns a tuple describing the frame:
[0] ... known+unknown symbols and scrambling
[1] ... modulation type after descrambling
[2] ... a boolean indicating whethere or not raw IQ samples needed"""
print('-------------------- get_frame --------------------', self._frame_counter)
return [self._preamble,self.MODE_BPSK,True] if self.is_preamble() else [self._data,self._mode,False]
def get_doppler(self, sy, sa):
"""used for doppler shift update, for determining which frame to provide next,
and for stopping at end of data/when the signal quality is too low
sy ... equalized symbols; sa ... samples"""
print('-------------------- get_doppler --------------------',self._counter,len(sy),len(sa))
success,doppler = self.quality_preamble(sy,sa) if self._counter == 0 else self.quality_data(sy)
self._counter = (self._counter+1)&1 if success else 0
self._is_first_frame = not success
def get_doppler(self, symbols, iq_samples):
"""returns a tuple
[0] ... quality flag
[1] ... doppler estimate (rad/symbol) if available"""
print('-------------------- get_doppler --------------------', self._frame_counter,len(symbols),len(iq_samples))
success,doppler = self.quality_preamble(symbols,iq_samples) if self.is_preamble() else self.quality_data(symbols)
if len(symbols) != 0:
self._frame_counter = (self._frame_counter+1)&1 if success else 0
self._is_first_frame = not success
return success,doppler
def quality_preamble(self, sy, sa):
sps = 5
zp = [x for x in self._preamble[0]['symb'][9:40] for i in range(sps)]
cc = np.array([np.sum(sa[ i*5:(31+i)*5]*zp) for i in range(49)])
imax = np.argmax(np.abs(cc[0:18]))
pks = cc[(imax,imax+15,imax+16,imax+31),]
apks = np.abs(pks)
test = np.mean(apks[(0,3),]) > 2*np.mean(apks[(1,2),])
doppler = np.diff(np.unwrap(np.angle(pks[(0,3),])))[0]/31 if test else 0
idx = range(80)
if self._is_first_frame:
idx = range(30,80)
z = sy[idx]*np.conj(self._preamble[0]['symb'][idx])
success = np.sum(np.real(z)<0) < 30
def is_preamble(self):
return self._frame_counter == 0
def quality_preamble(self, symbols, iq_samples):
"""quality check and doppler estimation for preamble"""
success = True
doppler = 0
if len(iq_samples) != 0:
zp = [x for x in self._preamble['symb'][9:40] for i in range(self._sps)]
cc = np.array([np.sum(iq_samples[ i*5:(31+i)*5]*zp) for i in range(49)])
imax = np.argmax(np.abs(cc[0:18]))
pks = cc[(imax,imax+15,imax+16,imax+31),]
apks = np.abs(pks)
success = np.mean(apks[(0,3),]) > 2*np.mean(apks[(1,2),])
doppler = np.diff(np.unwrap(np.angle(pks[(0,3),])))[0]/31 if success else 0
if len(symbols) != 0:
idx = range(30,80) if self._is_first_frame else range(80)
z = symbols[idx]*np.conj(self._preamble['symb'][idx])
success = np.sum(np.real(z)<0) < 30
return success,doppler
def quality_data(self, s):
"""quality check for the data frame"""
known_symbols = np.mod(range(176),48)>=32
success = np.sum(np.real(s[known_symbols])<0) < 20
return success,0 ## no doppler estimate for data frames
@ -87,7 +103,7 @@ class PhysicalLayer(object):
for j in range(3):
state = np.concatenate(([np.sum(state&taps)&1], state[0:-1]))
a=np.zeros(176, dtype=[('symb',np.complex64), ('scramble', np.complex64)])
## PSK-8 modulation
## 8PSK modulation
constellation = PhysicalLayer.make_psk(8,range(8))['points']
a['scramble'] = constellation[p,]
known_symbols = np.mod(range(176),48)>=32
@ -96,11 +112,13 @@ class PhysicalLayer(object):
@staticmethod
def make_psk(n, gray_code):
"""generates n-PSK constellation data"""
c = np.zeros(n, dtype=[('points', np.complex64), ('symbols', np.uint8)])
c['points'] = np.exp(2*np.pi*1j*np.array(range(n))/n)
c['symbols'] = gray_code
return c
## for now not used (doppler estimation after adaptive filtering does not work)
@staticmethod
def data_aided_frequency_estimation(x,c):
"""Data-Aided Frequency Estimation for Burst Digital Transmission,