259 lines
No EOL
8.1 KiB
C++
259 lines
No EOL
8.1 KiB
C++
#include <cstdlib>
|
|
#include <cstring>
|
|
#include <cstdio>
|
|
#include <cmath>
|
|
|
|
#include "common/wave.h"
|
|
#include "ft8/pack.h"
|
|
#include "ft8/encode.h"
|
|
#include "ft8/pack_v2.h"
|
|
#include "ft8/encode_v2.h"
|
|
|
|
#include "ft8/ldpc.h"
|
|
#include "fft/kiss_fftr.h"
|
|
|
|
|
|
void usage() {
|
|
printf("Decode a 15-second WAV file.\n");
|
|
}
|
|
|
|
|
|
float hann_i(int i, int N) {
|
|
float x = sinf((float)M_PI * i / (N - 1));
|
|
return x*x;
|
|
}
|
|
|
|
|
|
struct Candidate {
|
|
int16_t score;
|
|
uint16_t time_offset;
|
|
uint16_t freq_offset;
|
|
uint8_t time_sub;
|
|
uint8_t freq_sub;
|
|
};
|
|
|
|
|
|
void heapify_down(Candidate * heap, int heap_size) {
|
|
// heapify from the root down
|
|
int current = 0;
|
|
while (true) {
|
|
int largest = current;
|
|
int left = 2 * current + 1;
|
|
int right = left + 1;
|
|
|
|
if (left < heap_size && heap[left].score < heap[largest].score) {
|
|
largest = left;
|
|
}
|
|
if (right < heap_size && heap[right].score < heap[largest].score) {
|
|
largest = right;
|
|
}
|
|
if (largest == current) {
|
|
break;
|
|
}
|
|
|
|
Candidate tmp = heap[largest];
|
|
heap[largest] = heap[current];
|
|
heap[current] = tmp;
|
|
current = largest;
|
|
}
|
|
}
|
|
|
|
|
|
void heapify_up(Candidate * heap, int heap_size) {
|
|
// heapify from the last node up
|
|
int current = heap_size - 1;
|
|
while (current > 0) {
|
|
int parent = (current - 1) / 2;
|
|
if (heap[current].score >= heap[parent].score) {
|
|
break;
|
|
}
|
|
|
|
Candidate tmp = heap[parent];
|
|
heap[parent] = heap[current];
|
|
heap[current] = tmp;
|
|
current = parent;
|
|
}
|
|
}
|
|
|
|
|
|
// Find top N candidates in frequency and time according to their sync strength (looking at Costas symbols)
|
|
void find_sync(const uint8_t * power, int num_blocks, int num_bins, int num_candidates, Candidate * heap) {
|
|
// Costas 7x7 tone pattern
|
|
const uint8_t ICOS7[] = { 2,5,6,0,4,1,3 };
|
|
|
|
int heap_size = 0;
|
|
|
|
for (int alt = 0; alt < 4; ++alt) {
|
|
for (int time_offset = 0; time_offset < num_blocks - NN; ++time_offset) {
|
|
for (int freq_offset = 0; freq_offset < num_bins - 8; ++freq_offset) {
|
|
int score = 0;
|
|
|
|
// Compute score over bins 0-7, 36-43, 72-79
|
|
for (int m = 0; m <= 72; m += 36) {
|
|
for (int k = 0; k < 7; ++k) {
|
|
int offset = ((time_offset + k + m) * 4 + alt) * num_bins + freq_offset;
|
|
// score += 8 * (int)power[time_offset + k + m][alt][freq_offset + ICOS7[k]] -
|
|
score += 8 * (int)power[offset + ICOS7[k]] -
|
|
power[offset + 0] - power[offset + 1] -
|
|
power[offset + 2] - power[offset + 3] -
|
|
power[offset + 4] - power[offset + 5] -
|
|
power[offset + 6] - power[offset + 7];
|
|
}
|
|
}
|
|
|
|
// update the candidate list
|
|
if (heap_size == num_candidates && score > heap[0].score) {
|
|
// extract the least promising candidate
|
|
heap[0] = heap[heap_size - 1];
|
|
--heap_size;
|
|
|
|
heapify_down(heap, heap_size);
|
|
}
|
|
|
|
if (heap_size < num_candidates) {
|
|
// add the current candidate
|
|
heap[heap_size].score = score;
|
|
heap[heap_size].time_offset = time_offset;
|
|
heap[heap_size].freq_offset = freq_offset;
|
|
heap[heap_size].time_sub = alt / 2;
|
|
heap[heap_size].freq_sub = alt % 2;
|
|
++heap_size;
|
|
|
|
heapify_up(heap, heap_size);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
|
|
// Compute FFT magnitudes (log power) for each timeslot in the signal
|
|
void extract_power(const float * signal, int num_blocks, int num_bins, uint8_t * power) {
|
|
const int block_size = 2 * num_bins; // Average over 2 bins per FSK tone
|
|
const int nfft = 2 * block_size; // We take FFT of two blocks, advancing by one
|
|
|
|
float window[nfft];
|
|
for (int i = 0; i < nfft; ++i) {
|
|
window[i] = hann_i(i, nfft);
|
|
}
|
|
|
|
size_t fft_work_size;
|
|
kiss_fftr_alloc(nfft, 0, 0, &fft_work_size);
|
|
|
|
printf("N_FFT = %d\n", nfft);
|
|
printf("FFT work area = %lu\n", fft_work_size);
|
|
|
|
void * fft_work = malloc(fft_work_size);
|
|
kiss_fftr_cfg fft_cfg = kiss_fftr_alloc(nfft, 0, fft_work, &fft_work_size);
|
|
|
|
int offset = 0;
|
|
float fft_norm = 1.0f / nfft;
|
|
for (int i = 0; i < num_blocks; ++i) {
|
|
// Loop over two possible time offsets (0 and block_size/2)
|
|
for (int time_sub = 0; time_sub <= block_size/2; time_sub += block_size/2) {
|
|
kiss_fft_scalar timedata[nfft];
|
|
kiss_fft_cpx freqdata[nfft/2 + 1];
|
|
float mag_db[nfft/2 + 1];
|
|
|
|
// Extract windowed signal block
|
|
for (int j = 0; j < nfft; ++j) {
|
|
timedata[j] = window[j] * signal[(i * block_size) + (j + time_sub)];
|
|
}
|
|
|
|
kiss_fftr(fft_cfg, timedata, freqdata);
|
|
|
|
// Compute log magnitude in decibels
|
|
for (int j = 0; j < nfft/2 + 1; ++j) {
|
|
float mag2 = fft_norm * (freqdata[j].i * freqdata[j].i + freqdata[j].r * freqdata[j].r);
|
|
mag_db[j] = 10.0f * log10f(1.0E-10f + mag2);
|
|
}
|
|
|
|
// Loop over two possible frequency bin offsets (for averaging)
|
|
for (int freq_sub = 0; freq_sub < 2; ++freq_sub) {
|
|
for (int j = 0; j < num_bins; ++j) {
|
|
float db1 = mag_db[j * 2 + freq_sub];
|
|
float db2 = mag_db[j * 2 + freq_sub + 1];
|
|
float db = (db1 + db2) / 2;
|
|
|
|
// Scale decibels to unsigned 8-bit range
|
|
int scaled = (int)(0.5f + 2 * (db + 100));
|
|
power[offset] = (scaled < 0) ? 0 : ((scaled > 255) ? 255 : scaled);
|
|
++offset;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
free(fft_work);
|
|
}
|
|
|
|
|
|
int main(int argc, char ** argv) {
|
|
// Expect one command-line argument
|
|
if (argc < 2) {
|
|
usage();
|
|
return -1;
|
|
}
|
|
|
|
const char * wav_path = argv[1];
|
|
|
|
int sample_rate = 12000;
|
|
int num_samples = 15 * sample_rate;
|
|
float signal[num_samples];
|
|
|
|
int rc = load_wav(signal, num_samples, sample_rate, wav_path);
|
|
if (rc < 0) {
|
|
return -1;
|
|
}
|
|
|
|
const float fsk_dev = 6.25f;
|
|
|
|
const int num_bins = (int)(sample_rate / (2 * fsk_dev));
|
|
const int block_size = 2 * num_bins;
|
|
const int num_blocks = (num_samples - (block_size/2) - block_size) / block_size;
|
|
uint8_t power[num_blocks * 4 * num_bins]; // [num_blocks][4][num_bins] ~ 200 KB
|
|
|
|
printf("%d blocks, %d bins\n", num_blocks, num_bins);
|
|
|
|
extract_power(signal, num_blocks, num_bins, power);
|
|
|
|
const int num_candidates = 250;
|
|
Candidate heap[num_candidates];
|
|
|
|
find_sync(power, num_blocks, num_bins, num_candidates, heap);
|
|
|
|
for (int i = 0; i < num_candidates; ++i) {
|
|
float freq_offset = (heap[i].freq_offset + heap[i].freq_sub / 2.0f) * fsk_dev;
|
|
float time_offset = (heap[i].time_offset + heap[i].time_sub / 2.0f) / fsk_dev;
|
|
// int offset = (heap[i].time_offset * 4 + heap[i].time_sub * 2 + heap[i].freq_sub) * num_bins + heap[i].freq_offset;
|
|
printf("%03d: score = %.1f freq = %.1f time = %.2f\n", i, heap[i].score / 7.0f / 2, freq_offset, time_offset);
|
|
}
|
|
|
|
/*
|
|
|
|
// take absolute magnitude
|
|
s2(0:7,k)=abs(csymb(1:8))/1e3
|
|
|
|
// skip Costas sync symbols
|
|
s1(0:7,j)=s2(0:7,k)
|
|
|
|
// Normalize by median magnitude
|
|
s1=s1/xmeds1
|
|
|
|
// Extract bit significance
|
|
ps=s1(0:7,j)
|
|
bmeta(i4)=max(ps(4),ps(5),ps(6),ps(7))-max(ps(0),ps(1),ps(2),ps(3))
|
|
bmeta(i2)=max(ps(2),ps(3),ps(6),ps(7))-max(ps(0),ps(1),ps(4),ps(5))
|
|
bmeta(i1)=max(ps(1),ps(3),ps(5),ps(7))-max(ps(0),ps(2),ps(4),ps(6))
|
|
|
|
// Normalize by std. deviation
|
|
call normalizebmet(bmeta,3*ND)
|
|
|
|
// Magical fudge/scale factor
|
|
scalefac=2.83
|
|
llr0=scalefac*bmeta
|
|
|
|
*/
|
|
|
|
return 0;
|
|
} |