mirrors/gnss-sdr
1
0
Fork 0
mirror of https://github.com/gnss-sdr/gnss-sdr synced 2025-11-01 07:43:04 +00:00
Code Issues Releases Wiki Activity

Merge branch 'next' of https://github.com/mmajoral/gnss-sdr into fpga_extended_coherent_integration

Browse Source
This commit is contained in:
Marc Majoral
2019-07-18 11:35:34 +02:00
parent 4c22a1ce1a 76222945bd
commit dd996bc2e2
115 changed files with 2278 additions and 2030 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
src/algorithms/acquisition/adapters/galileo_e1_pcps_8ms_ambiguous_acquisition.cc
View File

@@ -203,7 +203,6 @@ signed int GalileoE1Pcps8msAmbiguousAcquisition::mag()
void GalileoE1Pcps8msAmbiguousAcquisition::init()
{
acquisition_cc_->init();
//set_local_code();
}
@@ -215,8 +214,10 @@ void GalileoE1Pcps8msAmbiguousAcquisition::set_local_code()
"Acquisition" + std::to_string(channel_) + ".cboc", false);
std::unique_ptr<std::complex<float>> code{new std::complex<float>[code_length_]};
std::array<char, 3> Signal_;
std::memcpy(Signal_.data(), gnss_synchro_->Signal, 3);
std::array<char, 3> Signal_{};
Signal_[0] = gnss_synchro_->Signal[0];
Signal_[1] = gnss_synchro_->Signal[1];
Signal_[2] = '\0';
galileo_e1_code_gen_complex_sampled(gsl::span<std::complex<float>>(code, code_length_), Signal_,
cboc, gnss_synchro_->PRN, fs_in_, 0, false);

9
src/algorithms/acquisition/adapters/galileo_e1_pcps_ambiguous_acquisition.cc
View File

@@ -232,7 +232,6 @@ signed int GalileoE1PcpsAmbiguousAcquisition::mag()
void GalileoE1PcpsAmbiguousAcquisition::init()
{
acquisition_->init();
//set_local_code();
}
@@ -246,7 +245,7 @@ void GalileoE1PcpsAmbiguousAcquisition::set_local_code()
if (acquire_pilot_ == true)
{
//set local signal generator to Galileo E1 pilot component (1C)
// set local signal generator to Galileo E1 pilot component (1C)
std::array<char, 3> pilot_signal = {{'1', 'C', '\0'}};
if (acq_parameters_.use_automatic_resampler)
{
@@ -261,8 +260,10 @@ void GalileoE1PcpsAmbiguousAcquisition::set_local_code()
}
else
{
std::array<char, 3> Signal_;
std::memcpy(Signal_.data(), gnss_synchro_->Signal, 3);
std::array<char, 3> Signal_{};
Signal_[0] = gnss_synchro_->Signal[0];
Signal_[1] = gnss_synchro_->Signal[1];
Signal_[2] = '\0';
if (acq_parameters_.use_automatic_resampler)
{
galileo_e1_code_gen_complex_sampled(code_span, Signal_,

9
src/algorithms/acquisition/adapters/galileo_e1_pcps_quicksync_ambiguous_acquisition.cc
View File

@@ -76,7 +76,7 @@ GalileoE1PcpsQuickSyncAmbiguousAcquisition::GalileoE1PcpsQuickSyncAmbiguousAcqui
This may be a bug, but acquisition also work by variying the folding factor at va-
lues different that the expressed in the paper. In adition, it is important to point
out that by making the folding factor smaller we were able to get QuickSync work with
Galileo. Future work should be directed to test this asumption statistically.*/
Galileo. Future work should be directed to test this assumption statistically.*/
//folding_factor_ = static_cast<unsigned int>(ceil(sqrt(log2(code_length_))));
folding_factor_ = configuration_->property(role + ".folding_factor", 2);
@@ -237,7 +237,6 @@ GalileoE1PcpsQuickSyncAmbiguousAcquisition::mag()
void GalileoE1PcpsQuickSyncAmbiguousAcquisition::init()
{
acquisition_cc_->init();
//set_local_code();
}
@@ -249,8 +248,10 @@ void GalileoE1PcpsQuickSyncAmbiguousAcquisition::set_local_code()
"Acquisition" + std::to_string(channel_) + ".cboc", false);
std::unique_ptr<std::complex<float>> code{new std::complex<float>[code_length_]};
std::array<char, 3> Signal_;
std::memcpy(Signal_.data(), gnss_synchro_->Signal, 3);
std::array<char, 3> Signal_{};
Signal_[0] = gnss_synchro_->Signal[0];
Signal_[1] = gnss_synchro_->Signal[1];
Signal_[2] = '\0';
galileo_e1_code_gen_complex_sampled(gsl::span<std::complex<float>>(code.get(), code_length_), Signal_,
cboc, gnss_synchro_->PRN, fs_in_, 0, false);

7
src/algorithms/acquisition/adapters/galileo_e1_pcps_tong_ambiguous_acquisition.cc
View File

@@ -207,7 +207,6 @@ signed int GalileoE1PcpsTongAmbiguousAcquisition::mag()
void GalileoE1PcpsTongAmbiguousAcquisition::init()
{
acquisition_cc_->init();
//set_local_code();
}
@@ -219,8 +218,10 @@ void GalileoE1PcpsTongAmbiguousAcquisition::set_local_code()
"Acquisition" + std::to_string(channel_) + ".cboc", false);
std::unique_ptr<std::complex<float>> code{new std::complex<float>[code_length_]};
std::array<char, 3> Signal_;
std::memcpy(Signal_.data(), gnss_synchro_->Signal, 3);
std::array<char, 3> Signal_{};
Signal_[0] = gnss_synchro_->Signal[0];
Signal_[1] = gnss_synchro_->Signal[1];
Signal_[2] = '\0';
galileo_e1_code_gen_complex_sampled(gsl::span<std::complex<float>>(code.get(), code_length_), Signal_,
cboc, gnss_synchro_->PRN, fs_in_, 0, false);

2
src/algorithms/acquisition/adapters/galileo_e5a_pcps_acquisition.cc
View File

@@ -234,7 +234,7 @@ void GalileoE5aPcpsAcquisition::init()
void GalileoE5aPcpsAcquisition::set_local_code()
{
std::unique_ptr<std::complex<float>> code{new std::complex<float>[code_length_]};
std::array<char, 3> signal_;
std::array<char, 3> signal_{};
signal_[0] = '5';
signal_[2] = '\0';

37
src/algorithms/acquisition/gnuradio_blocks/pcps_acquisition.cc
View File

@@ -56,8 +56,9 @@
#include <volk/volk.h>
#include <volk_gnsssdr/volk_gnsssdr.h>
#include <algorithm> // for fill_n, min
#include <cmath> // for floor, fmod, rint, ceil
#include <cstring> // for memcpy
#include <array>
#include <cmath> // for floor, fmod, rint, ceil
#include <cstring> // for memcpy
#include <iostream>
#include <map>
@@ -455,56 +456,56 @@ void pcps_acquisition::dump_results(int32_t effective_fft_size)
}
else
{
size_t dims[2] = {static_cast<size_t>(effective_fft_size), static_cast<size_t>(d_num_doppler_bins)};
matvar_t* matvar = Mat_VarCreate("acq_grid", MAT_C_SINGLE, MAT_T_SINGLE, 2, dims, grid_.memptr(), 0);
std::array<size_t, 2> dims{static_cast<size_t>(effective_fft_size), static_cast<size_t>(d_num_doppler_bins)};
matvar_t* matvar = Mat_VarCreate("acq_grid", MAT_C_SINGLE, MAT_T_SINGLE, 2, dims.data(), grid_.memptr(), 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
dims[0] = static_cast<size_t>(1);
dims[1] = static_cast<size_t>(1);
matvar = Mat_VarCreate("doppler_max", MAT_C_INT32, MAT_T_INT32, 1, dims, &acq_parameters.doppler_max, 0);
matvar = Mat_VarCreate("doppler_max", MAT_C_INT32, MAT_T_INT32, 1, dims.data(), &acq_parameters.doppler_max, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("doppler_step", MAT_C_INT32, MAT_T_INT32, 1, dims, &d_doppler_step, 0);
matvar = Mat_VarCreate("doppler_step", MAT_C_INT32, MAT_T_INT32, 1, dims.data(), &d_doppler_step, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("d_positive_acq", MAT_C_INT32, MAT_T_INT32, 1, dims, &d_positive_acq, 0);
matvar = Mat_VarCreate("d_positive_acq", MAT_C_INT32, MAT_T_INT32, 1, dims.data(), &d_positive_acq, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
auto aux = static_cast<float>(d_gnss_synchro->Acq_doppler_hz);
matvar = Mat_VarCreate("acq_doppler_hz", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims, &aux, 0);
matvar = Mat_VarCreate("acq_doppler_hz", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims.data(), &aux, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
aux = static_cast<float>(d_gnss_synchro->Acq_delay_samples);
matvar = Mat_VarCreate("acq_delay_samples", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims, &aux, 0);
matvar = Mat_VarCreate("acq_delay_samples", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims.data(), &aux, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("test_statistic", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims, &d_test_statistics, 0);
matvar = Mat_VarCreate("test_statistic", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims.data(), &d_test_statistics, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("threshold", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims, &d_threshold, 0);
matvar = Mat_VarCreate("threshold", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims.data(), &d_threshold, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("input_power", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims, &d_input_power, 0);
matvar = Mat_VarCreate("input_power", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims.data(), &d_input_power, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("sample_counter", MAT_C_UINT64, MAT_T_UINT64, 1, dims, &d_sample_counter, 0);
matvar = Mat_VarCreate("sample_counter", MAT_C_UINT64, MAT_T_UINT64, 1, dims.data(), &d_sample_counter, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("PRN", MAT_C_UINT32, MAT_T_UINT32, 1, dims, &d_gnss_synchro->PRN, 0);
matvar = Mat_VarCreate("PRN", MAT_C_UINT32, MAT_T_UINT32, 1, dims.data(), &d_gnss_synchro->PRN, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("num_dwells", MAT_C_INT32, MAT_T_INT32, 1, dims, &d_num_noncoherent_integrations_counter, 0);
matvar = Mat_VarCreate("num_dwells", MAT_C_INT32, MAT_T_INT32, 1, dims.data(), &d_num_noncoherent_integrations_counter, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
@@ -512,18 +513,18 @@ void pcps_acquisition::dump_results(int32_t effective_fft_size)
{
dims[0] = static_cast<size_t>(effective_fft_size);
dims[1] = static_cast<size_t>(d_num_doppler_bins_step2);
matvar = Mat_VarCreate("acq_grid_narrow", MAT_C_SINGLE, MAT_T_SINGLE, 2, dims, narrow_grid_.memptr(), 0);
matvar = Mat_VarCreate("acq_grid_narrow", MAT_C_SINGLE, MAT_T_SINGLE, 2, dims.data(), narrow_grid_.memptr(), 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
dims[0] = static_cast<size_t>(1);
dims[1] = static_cast<size_t>(1);
matvar = Mat_VarCreate("doppler_step_narrow", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims, &acq_parameters.doppler_step2, 0);
matvar = Mat_VarCreate("doppler_step_narrow", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims.data(), &acq_parameters.doppler_step2, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
aux = d_doppler_center_step_two - static_cast<float>(floor(d_num_doppler_bins_step2 / 2.0)) * acq_parameters.doppler_step2;
matvar = Mat_VarCreate("doppler_grid_narrow_min", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims, &aux, 0);
matvar = Mat_VarCreate("doppler_grid_narrow_min", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims.data(), &aux, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
}

27
src/algorithms/acquisition/gnuradio_blocks/pcps_acquisition_fine_doppler_cc.cc
View File

@@ -49,6 +49,7 @@
#include <volk/volk.h>
#include <volk_gnsssdr/volk_gnsssdr.h>
#include <algorithm> // std::rotate, std::fill_n
#include <array>
#include <sstream>
#if HAS_STD_FILESYSTEM
@@ -512,7 +513,7 @@ int pcps_acquisition_fine_doppler_cc::general_work(int noutput_items,
{
/*!
* TODO: High sensitivity acquisition algorithm:
* State Mechine:
* State Machine:
* S0. StandBy. If d_active==1 -> S1
* S1. ComputeGrid. Perform the FFT acqusition doppler and delay grid.
* Accumulate the search grid matrix (#doppler_bins x #fft_size)
@@ -673,52 +674,52 @@ void pcps_acquisition_fine_doppler_cc::dump_results(int effective_fft_size)
}
else
{
size_t dims[2] = {static_cast<size_t>(effective_fft_size), static_cast<size_t>(d_num_doppler_points)};
matvar_t *matvar = Mat_VarCreate("acq_grid", MAT_C_SINGLE, MAT_T_SINGLE, 2, dims, grid_.memptr(), 0);
std::array<size_t, 2> dims{static_cast<size_t>(effective_fft_size), static_cast<size_t>(d_num_doppler_points)};
matvar_t *matvar = Mat_VarCreate("acq_grid", MAT_C_SINGLE, MAT_T_SINGLE, 2, dims.data(), grid_.memptr(), 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
dims[0] = static_cast<size_t>(1);
dims[1] = static_cast<size_t>(1);
matvar = Mat_VarCreate("doppler_max", MAT_C_INT32, MAT_T_INT32, 1, dims, &d_config_doppler_max, 0);
matvar = Mat_VarCreate("doppler_max", MAT_C_INT32, MAT_T_INT32, 1, dims.data(), &d_config_doppler_max, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("doppler_step", MAT_C_INT32, MAT_T_INT32, 1, dims, &d_doppler_step, 0);
matvar = Mat_VarCreate("doppler_step", MAT_C_INT32, MAT_T_INT32, 1, dims.data(), &d_doppler_step, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("d_positive_acq", MAT_C_INT32, MAT_T_INT32, 1, dims, &d_positive_acq, 0);
matvar = Mat_VarCreate("d_positive_acq", MAT_C_INT32, MAT_T_INT32, 1, dims.data(), &d_positive_acq, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
auto aux = static_cast<float>(d_gnss_synchro->Acq_doppler_hz);
matvar = Mat_VarCreate("acq_doppler_hz", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims, &aux, 0);
matvar = Mat_VarCreate("acq_doppler_hz", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims.data(), &aux, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
aux = static_cast<float>(d_gnss_synchro->Acq_delay_samples);
matvar = Mat_VarCreate("acq_delay_samples", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims, &aux, 0);
matvar = Mat_VarCreate("acq_delay_samples", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims.data(), &aux, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("test_statistic", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims, &d_test_statistics, 0);
matvar = Mat_VarCreate("test_statistic", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims.data(), &d_test_statistics, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("threshold", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims, &d_threshold, 0);
matvar = Mat_VarCreate("threshold", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims.data(), &d_threshold, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
aux = 0.0;
matvar = Mat_VarCreate("input_power", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims, &aux, 0);
matvar = Mat_VarCreate("input_power", MAT_C_SINGLE, MAT_T_SINGLE, 1, dims.data(), &aux, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("sample_counter", MAT_C_UINT64, MAT_T_UINT64, 1, dims, &d_sample_counter, 0);
matvar = Mat_VarCreate("sample_counter", MAT_C_UINT64, MAT_T_UINT64, 1, dims.data(), &d_sample_counter, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);
matvar = Mat_VarCreate("PRN", MAT_C_UINT32, MAT_T_UINT32, 1, dims, &d_gnss_synchro->PRN, 0);
matvar = Mat_VarCreate("PRN", MAT_C_UINT32, MAT_T_UINT32, 1, dims.data(), &d_gnss_synchro->PRN, 0);
Mat_VarWrite(matfp, matvar, MAT_COMPRESSION_ZLIB); // or MAT_COMPRESSION_NONE
Mat_VarFree(matvar);

2
src/algorithms/acquisition/gnuradio_blocks/pcps_acquisition_fpga.h
View File

@@ -182,7 +182,7 @@ public:
}
/*!
* \brief This funciton triggers a HW reset of the FPGA PL.
* \brief This function triggers a HW reset of the FPGA PL.
*/
void reset_acquisition(void);

2
src/algorithms/acquisition/gnuradio_blocks/pcps_assisted_acquisition_cc.cc
View File

@@ -353,7 +353,7 @@ int pcps_assisted_acquisition_cc::general_work(int noutput_items,
{
/*!
* TODO: High sensitivity acquisition algorithm:
* State Mechine:
* State Machine:
* S0. StandBy. If d_active==1 -> S1
* S1. GetAssist. Define search grid with assistance information. Reset grid matrix -> S2
* S2. ComputeGrid. Perform the FFT acqusition doppler and delay grid.

114
src/algorithms/acquisition/gnuradio_blocks/pcps_quicksync_acquisition_cc.cc
View File

@@ -34,6 +34,7 @@
#include <gnuradio/io_signature.h>
#include <volk/volk.h>
#include <volk_gnsssdr/volk_gnsssdr.h>
#include <array>
#include <cmath>
#include <exception>
#include <sstream>
@@ -311,26 +312,26 @@ int pcps_quicksync_acquisition_cc::general_work(int noutput_items,
case 1:
{
/* initialize acquisition implementing the QuickSync algorithm*/
// initialize acquisition implementing the QuickSync algorithm
//DLOG(INFO) << "START CASE 1";
int32_t doppler;
uint32_t indext = 0;
float magt = 0.0;
const auto* in = reinterpret_cast<const gr_complex*>(input_items[0]); //Get the input samples pointer
const auto* in = reinterpret_cast<const gr_complex*>(input_items[0]); // Get the input samples pointer
auto* in_temp = static_cast<gr_complex*>(volk_gnsssdr_malloc(d_samples_per_code * d_folding_factor * sizeof(gr_complex), volk_gnsssdr_get_alignment()));
auto* in_temp_folded = static_cast<gr_complex*>(volk_gnsssdr_malloc(d_fft_size * sizeof(gr_complex), volk_gnsssdr_get_alignment()));
/*Create a signal to store a signal of size 1ms, to perform correlation
in time. No folding on this data is required*/
// Create a signal to store a signal of size 1ms, to perform correlation
// in time. No folding on this data is required
auto* in_1code = static_cast<gr_complex*>(volk_gnsssdr_malloc(d_samples_per_code * sizeof(gr_complex), volk_gnsssdr_get_alignment()));
/*Stores the values of the correlation output between the local code
and the signal with doppler shift corrected */
// Stores the values of the correlation output between the local code
// and the signal with doppler shift corrected
auto* corr_output = static_cast<gr_complex*>(volk_gnsssdr_malloc(d_samples_per_code * sizeof(gr_complex), volk_gnsssdr_get_alignment()));
/*Stores a copy of the folded version of the signal.This is used for
the FFT operations in future steps of execution*/
// Stores a copy of the folded version of the signal.This is used for
// the FFT operations in future steps of execution*/
// gr_complex in_folded[d_fft_size];
float fft_normalization_factor = static_cast<float>(d_fft_size) * static_cast<float>(d_fft_size);
@@ -354,35 +355,34 @@ int pcps_quicksync_acquisition_cc::general_work(int noutput_items,
<< d_samples_per_code * d_folding_factor;
/* 1- Compute the input signal power estimation. This operation is
being performed in a signal of size nxp */
// 1- Compute the input signal power estimation. This operation is
// being performed in a signal of size nxp
volk_32fc_magnitude_squared_32f(d_magnitude, in, d_samples_per_code * d_folding_factor);
volk_32f_accumulator_s32f(&d_input_power, d_magnitude, d_samples_per_code * d_folding_factor);
d_input_power /= static_cast<float>(d_samples_per_code * d_folding_factor);
for (uint32_t doppler_index = 0; doppler_index < d_num_doppler_bins; doppler_index++)
{
/*Ensure that the signal is going to start with all samples
at zero. This is done to avoid over acumulation when performing
the folding process to be stored in d_fft_if->get_inbuf()*/
// Ensure that the signal is going to start with all samples
// at zero. This is done to avoid over acumulation when performing
// the folding process to be stored in d_fft_if->get_inbuf()
d_signal_folded = new gr_complex[d_fft_size]();
memcpy(d_fft_if->get_inbuf(), d_signal_folded, sizeof(gr_complex) * (d_fft_size));
/*Doppler search steps and then multiplication of the incoming
signal with the doppler wipeoffs to eliminate frequency offset
*/
// Doppler search steps and then multiplication of the incoming
// signal with the doppler wipeoffs to eliminate frequency offset
doppler = -static_cast<int32_t>(d_doppler_max) + d_doppler_step * doppler_index;
/*Perform multiplication of the incoming signal with the
complex exponential vector. This removes the frequency doppler
shift offset*/
// Perform multiplication of the incoming signal with the
// complex exponential vector. This removes the frequency doppler
// shift offset
volk_32fc_x2_multiply_32fc(in_temp, in,
d_grid_doppler_wipeoffs[doppler_index],
d_samples_per_code * d_folding_factor);
/*Perform folding of the carrier wiped-off incoming signal. Since
superlinear method is being used the folding factor in the
incoming raw data signal is of d_folding_factor^2*/
// Perform folding of the carrier wiped-off incoming signal. Since
// superlinear method is being used the folding factor in the
// incoming raw data signal is of d_folding_factor^2
for (int32_t i = 0; i < static_cast<int32_t>(d_folding_factor * d_folding_factor); i++)
{
std::transform((in_temp + i * d_fft_size),
@@ -392,28 +392,26 @@ int pcps_quicksync_acquisition_cc::general_work(int noutput_items,
std::plus<gr_complex>());
}
/* 3- Perform the FFT-based convolution (parallel time search)
Compute the FFT of the carrier wiped--off incoming signal*/
// 3- Perform the FFT-based convolution (parallel time search)
// Compute the FFT of the carrier wiped--off incoming signal
d_fft_if->execute();
/*Multiply carrier wiped--off, Fourier transformed incoming
signal with the local FFT'd code reference using SIMD
operations with VOLK library*/
// Multiply carrier wiped--off, Fourier transformed incoming
// signal with the local FFT'd code reference using SIMD
// operations with VOLK library
volk_32fc_x2_multiply_32fc(d_ifft->get_inbuf(),
d_fft_if->get_outbuf(), d_fft_codes, d_fft_size);
/* compute the inverse FFT of the aliased signal*/
// compute the inverse FFT of the aliased signal
d_ifft->execute();
/* Compute the magnitude and get the maximum value with its
index position*/
// Compute the magnitude and get the maximum value with its
// index position
volk_32fc_magnitude_squared_32f(d_magnitude_folded,
d_ifft->get_outbuf(), d_fft_size);
/* Normalize the maximum value to correct the scale factor
introduced by FFTW*/
//volk_32f_s32f_multiply_32f_a(d_magnitude_folded,d_magnitude_folded,
// (1 / (fft_normalization_factor * fft_normalization_factor)), d_fft_size);
// Normalize the maximum value to correct the scale factor
// introduced by FFTW
volk_gnsssdr_32f_index_max_32u(&indext, d_magnitude_folded, d_fft_size);
magt = d_magnitude_folded[indext] / (fft_normalization_factor * fft_normalization_factor);
@@ -425,19 +423,18 @@ int pcps_quicksync_acquisition_cc::general_work(int noutput_items,
{
d_mag = magt;
/* In case that d_bit_transition_flag = true, we compare the potentially
new maximum test statistics (d_mag/d_input_power) with the value in
d_test_statistics. When the second dwell is being processed, the value
of d_mag/d_input_power could be lower than d_test_statistics (i.e,
the maximum test statistics in the previous dwell is greater than
current d_mag/d_input_power). Note that d_test_statistics is not
restarted between consecutive dwells in multidwell operation.*/
// In case that d_bit_transition_flag = true, we compare the potentially
// new maximum test statistics (d_mag/d_input_power) with the value in
// d_test_statistics. When the second dwell is being processed, the value
// of d_mag/d_input_power could be lower than d_test_statistics (i.e,
// the maximum test statistics in the previous dwell is greater than
// current d_mag/d_input_power). Note that d_test_statistics is not
// restarted between consecutive dwells in multidwell operation.
if (d_test_statistics < (d_mag / d_input_power) || !d_bit_transition_flag)
{
uint32_t detected_delay_samples_folded = 0;
detected_delay_samples_folded = (indext % d_samples_per_code);
gr_complex complex_acumulator[100];
//gr_complex complex_acumulator[d_folding_factor];
std::array<gr_complex, 100> complex_acumulator{};
for (int32_t i = 0; i < static_cast<int32_t>(d_folding_factor); i++)
{
@@ -446,16 +443,16 @@ int pcps_quicksync_acquisition_cc::general_work(int noutput_items,
for (int32_t i = 0; i < static_cast<int32_t>(d_folding_factor); i++)
{
/*Copy a signal of 1 code length into suggested buffer.
The copied signal must have doppler effect corrected*/
// Copy a signal of 1 code length into suggested buffer.
// The copied signal must have doppler effect corrected*/
memcpy(in_1code, &in_temp[d_possible_delay[i]],
sizeof(gr_complex) * (d_samples_per_code));
/*Perform multiplication of the unmodified local
generated code with the incoming signal with doppler
effect corrected and accumulates its value. This
is indeed correlation in time for an specific value
of a shift*/
// Perform multiplication of the unmodified local
// generated code with the incoming signal with doppler
// effect corrected and accumulates its value. This
// is indeed correlation in time for an specific value
// of a shift
volk_32fc_x2_multiply_32fc(corr_output, in_1code, d_code, d_samples_per_code);
for (int32_t j = 0; j < d_samples_per_code; j++)
@@ -463,28 +460,27 @@ int pcps_quicksync_acquisition_cc::general_work(int noutput_items,
complex_acumulator[i] += (corr_output[j]);
}
}
/*Obtain maximun value of correlation given the possible delay selected */
volk_32fc_magnitude_squared_32f(d_corr_output_f, complex_acumulator, d_folding_factor);
// Obtain maximum value of correlation given the possible delay selected
volk_32fc_magnitude_squared_32f(d_corr_output_f, complex_acumulator.data(), d_folding_factor);
volk_gnsssdr_32f_index_max_32u(&indext, d_corr_output_f, d_folding_factor);
/*Now save the real code phase in the gnss_syncro block for use in other stages*/
// Now save the real code phase in the gnss_syncro block for use in other stages
d_gnss_synchro->Acq_delay_samples = static_cast<double>(d_possible_delay[indext]);
d_gnss_synchro->Acq_doppler_hz = static_cast<double>(doppler);
d_gnss_synchro->Acq_samplestamp_samples = d_sample_counter;
d_gnss_synchro->Acq_doppler_step = d_doppler_step;
/* 5- Compute the test statistics and compare to the threshold d_test_statistics = 2 * d_fft_size * d_mag / d_input_power;*/
// 5- Compute the test statistics and compare to the threshold d_test_statistics = 2 * d_fft_size * d_mag / d_input_power;
d_test_statistics = d_mag / d_input_power;
//delete complex_acumulator;
}
}
// Record results to file if required
if (d_dump)
{
/*Since QuickSYnc performs a folded correlation in frequency by means
of the FFT, it is essential to also keep the values obtained from the
possible delay to show how it is maximize*/
// Since QuickSYnc performs a folded correlation in frequency by means
// of the FFT, it is essential to also keep the values obtained from the
// possible delay to show how it is maximize
std::stringstream filename;
std::streamsize n = sizeof(float) * (d_fft_size); // complex file write
filename.str("");
@@ -492,7 +488,7 @@ int pcps_quicksync_acquisition_cc::general_work(int noutput_items,
<< "_" << d_gnss_synchro->Signal << "_sat_"
<< d_gnss_synchro->PRN << "_doppler_" << doppler << ".dat";
d_dump_file.open(filename.str().c_str(), std::ios::out | std::ios::binary);
d_dump_file.write(reinterpret_cast<char*>(d_magnitude_folded), n); //write directly |abs(x)|^2 in this Doppler bin?
d_dump_file.write(reinterpret_cast<char*>(d_magnitude_folded), n); // write directly |abs(x)|^2 in this Doppler bin?
d_dump_file.close();
}
}

2
src/algorithms/acquisition/libs/acq_conf.cc
View File

@@ -1,7 +1,7 @@
/*!
* \file acq_conf.cc
* \brief Class that contains all the configuration parameters for generic
* acquisition block based on the PCPS algoritm.
* acquisition block based on the PCPS algorithm.
* \author Carles Fernandez, 2018. cfernandez(at)cttc.es
*
* -------------------------------------------------------------------------

2
src/algorithms/acquisition/libs/acq_conf.h
View File

@@ -1,7 +1,7 @@
/*!
* \file acq_conf.h
* \brief Class that contains all the configuration parameters for generic
* acquisition block based on the PCPS algoritm.
* acquisition block based on the PCPS algorithm.
* \author Carles Fernandez, 2018. cfernandez(at)cttc.es
*
* -------------------------------------------------------------------------

2
src/algorithms/acquisition/libs/fpga_acquisition.h
View File

@@ -121,7 +121,7 @@ private:
int32_t d_fd; // driver descriptor
volatile uint32_t *d_map_base; // driver memory map
uint32_t *d_all_fft_codes; // memory that contains all the code ffts
uint32_t d_vector_length; // number of samples incluing padding and number of ms
uint32_t d_vector_length; // number of samples including padding and number of ms
uint32_t d_excludelimit;
uint32_t d_nsamples_total; // number of samples including padding
uint32_t d_nsamples; // number of samples not including padding