Skip to content

O
OpenXG-RAN
Commit 2d3ac580 authored by francescomani's avatar francescomani
Browse files
Options

formatting

parent ccdabb4c
Hide whitespace changes
Inline Side-by-side
Showing with 64 additions and 63 deletions
openair1/PHY/NR_UE_TRANSPORT/nr_dlsch_demodulation.c
View file @ 2d3ac580
...@@ -1257,7 +1257,7 @@ static void nr_dlsch_extract_rbs(uint32_t rxdataF_sz, ...@@ -1257,7 +1257,7 @@ static void nr_dlsch_extract_rbs(uint32_t rxdataF_sz,
} }
static void nr_dlsch_detection_mrc(uint32_t rx_size_symbol, static void nr_dlsch_detection_mrc(uint32_t rx_size_symbol,
short n_tx, short nl,
short n_rx, short n_rx,
int32_t rxdataF_comp[][n_rx][rx_size_symbol * NR_SYMBOLS_PER_SLOT], int32_t rxdataF_comp[][n_rx][rx_size_symbol * NR_SYMBOLS_PER_SLOT],
int ***rho, int ***rho,
...@@ -1268,19 +1268,19 @@ static void nr_dlsch_detection_mrc(uint32_t rx_size_symbol, ...@@ -1268,19 +1268,19 @@ static void nr_dlsch_detection_mrc(uint32_t rx_size_symbol,
int length) int length)
{ {
simde__m128i *rxdataF_comp128_0,*rxdataF_comp128_1,*dl_ch_mag128_0,*dl_ch_mag128_1,*dl_ch_mag128_0b,*dl_ch_mag128_1b,*dl_ch_mag128_0r,*dl_ch_mag128_1r; simde__m128i *rxdataF_comp128_0,*rxdataF_comp128_1,*dl_ch_mag128_0,*dl_ch_mag128_1,*dl_ch_mag128_0b,*dl_ch_mag128_1b,*dl_ch_mag128_0r,*dl_ch_mag128_1r;
uint32_t nb_rb_0 = length/12 + ((length%12)?1:0); uint32_t nb_rb_0 = length / 12 + ((length % 12) ? 1 : 0);
if (n_rx > 1) { if (n_rx > 1) {
for (int aatx = 0; aatx < n_tx; aatx++) { for (int l = 0; l < nl; l++) {
rxdataF_comp128_0 = (simde__m128i *)(rxdataF_comp[aatx][0] + symbol * rx_size_symbol); rxdataF_comp128_0 = (simde__m128i *)(rxdataF_comp[l][0] + symbol * rx_size_symbol);
dl_ch_mag128_0 = (simde__m128i *)dl_ch_mag[aatx][0]; dl_ch_mag128_0 = (simde__m128i *)dl_ch_mag[l][0];
dl_ch_mag128_0b = (simde__m128i *)dl_ch_magb[aatx][0]; dl_ch_mag128_0b = (simde__m128i *)dl_ch_magb[l][0];
dl_ch_mag128_0r = (simde__m128i *)dl_ch_magr[aatx][0]; dl_ch_mag128_0r = (simde__m128i *)dl_ch_magr[l][0];
for (int aarx = 1; aarx < n_rx; aarx++) { for (int aarx = 1; aarx < n_rx; aarx++) {
rxdataF_comp128_1 = (simde__m128i *)(rxdataF_comp[aatx][aarx] + symbol * rx_size_symbol); rxdataF_comp128_1 = (simde__m128i *)(rxdataF_comp[l][aarx] + symbol * rx_size_symbol);
dl_ch_mag128_1 = (simde__m128i *)dl_ch_mag[aatx][aarx]; dl_ch_mag128_1 = (simde__m128i *)dl_ch_mag[l][aarx];
dl_ch_mag128_1b = (simde__m128i *)dl_ch_magb[aatx][aarx]; dl_ch_mag128_1b = (simde__m128i *)dl_ch_magb[l][aarx];
dl_ch_mag128_1r = (simde__m128i *)dl_ch_magr[aatx][aarx]; dl_ch_mag128_1r = (simde__m128i *)dl_ch_magr[l][aarx];
// MRC on each re of rb, both on MF output and magnitude (for 16QAM/64QAM/256 llr computation) // MRC on each re of rb, both on MF output and magnitude (for 16QAM/64QAM/256 llr computation)
for (int i = 0; i < nb_rb_0 * 3; i++) { for (int i = 0; i < nb_rb_0 * 3; i++) {
...@@ -1293,7 +1293,7 @@ static void nr_dlsch_detection_mrc(uint32_t rx_size_symbol, ...@@ -1293,7 +1293,7 @@ static void nr_dlsch_detection_mrc(uint32_t rx_size_symbol,
} }
#ifdef DEBUG_DLSCH_DEMOD #ifdef DEBUG_DLSCH_DEMOD
for (int i = 0; i < nb_rb_0 * 3; i++) { for (int i = 0; i < nb_rb_0 * 3; i++) {
printf("symbol%d RB %d\n", symbol,i / 3); printf("symbol%d RB %d\n", symbol, i / 3);
rxdataF_comp128_0 = (simde__m128i *)(rxdataF_comp[0][0] + symbol * rx_size_symbol); rxdataF_comp128_0 = (simde__m128i *)(rxdataF_comp[0][0] + symbol * rx_size_symbol);
rxdataF_comp128_1 = (simde__m128i *)(rxdataF_comp[0][n_rx] + symbol * rx_size_symbol); rxdataF_comp128_1 = (simde__m128i *)(rxdataF_comp[0][n_rx] + symbol * rx_size_symbol);
print_shorts("tx 1 mrc_re/mrc_Im:",(int16_t*)&rxdataF_comp128_0[i]); print_shorts("tx 1 mrc_re/mrc_Im:",(int16_t*)&rxdataF_comp128_0[i]);
...@@ -1677,7 +1677,7 @@ void nr_conjch0_mult_ch1(int *ch0, ...@@ -1677,7 +1677,7 @@ void nr_conjch0_mult_ch1(int *ch0,
*/ */
static void nr_dlsch_mmse(uint32_t rx_size_symbol, static void nr_dlsch_mmse(uint32_t rx_size_symbol,
unsigned char n_rx, unsigned char n_rx,
unsigned char n_tx, // number of layer unsigned char nl, // number of layer
int32_t rxdataF_comp[][n_rx][rx_size_symbol * NR_SYMBOLS_PER_SLOT], int32_t rxdataF_comp[][n_rx][rx_size_symbol * NR_SYMBOLS_PER_SLOT],
int32_t dl_ch_mag[][n_rx][rx_size_symbol], int32_t dl_ch_mag[][n_rx][rx_size_symbol],
int32_t dl_ch_magb[][n_rx][rx_size_symbol], int32_t dl_ch_magb[][n_rx][rx_size_symbol],
...@@ -1690,25 +1690,24 @@ static void nr_dlsch_mmse(uint32_t rx_size_symbol, ...@@ -1690,25 +1690,24 @@ static void nr_dlsch_mmse(uint32_t rx_size_symbol,
int length, int length,
uint32_t noise_var) uint32_t noise_var)
{ {
int *ch0r, *ch0c; uint32_t nb_rb_0 = length / 12 + ((length % 12) ? 1 : 0);
uint32_t nb_rb_0 = length/12 + ((length%12)?1:0);
c16_t determ_fin[12 * nb_rb_0] __attribute__((aligned(32))); c16_t determ_fin[12 * nb_rb_0] __attribute__((aligned(32)));
///Allocate H^*H matrix elements and sub elements ///Allocate H^*H matrix elements and sub elements
c16_t conjH_H_elements_data[n_rx][n_tx][n_tx][12 * nb_rb_0]; c16_t conjH_H_elements_data[n_rx][nl][nl][12 * nb_rb_0];
memset(conjH_H_elements_data, 0, sizeof(conjH_H_elements_data)); memset(conjH_H_elements_data, 0, sizeof(conjH_H_elements_data));
c16_t *conjH_H_elements[n_rx][n_tx][n_tx]; c16_t *conjH_H_elements[n_rx][nl][nl];
for (int aarx = 0; aarx < n_rx; aarx++) for (int aarx = 0; aarx < n_rx; aarx++)
for (int rtx = 0; rtx < n_tx; rtx++) for (int rtx = 0; rtx < nl; rtx++)
for (int ctx = 0; ctx < n_tx; ctx++) for (int ctx = 0; ctx < nl; ctx++)
conjH_H_elements[aarx][rtx][ctx] = conjH_H_elements_data[aarx][rtx][ctx]; conjH_H_elements[aarx][rtx][ctx] = conjH_H_elements_data[aarx][rtx][ctx];
//Compute H^*H matrix elements and sub elements:(1/2^log2_maxh)*conjH_H_elements //Compute H^*H matrix elements and sub elements:(1/2^log2_maxh)*conjH_H_elements
for (int rtx=0;rtx<n_tx;rtx++) {//row for (int rtx = 0; rtx < nl; rtx++) {//row
for (int ctx=0;ctx<n_tx;ctx++) {//column for (int ctx = 0; ctx < nl; ctx++) {//column
for (int aarx=0;aarx<n_rx;aarx++) { for (int aarx = 0; aarx < n_rx; aarx++) {
ch0r = (int *)dl_ch_estimates_ext[rtx*n_rx+aarx]; int *ch0r = (int *)dl_ch_estimates_ext[rtx * n_rx + aarx];
ch0c = (int *)dl_ch_estimates_ext[ctx*n_rx+aarx]; int *ch0c = (int *)dl_ch_estimates_ext[ctx * n_rx + aarx];
nr_conjch0_mult_ch1(ch0r, nr_conjch0_mult_ch1(ch0r,
ch0c, ch0c,
(int32_t *)conjH_H_elements[aarx][ctx][rtx], // sic (int32_t *)conjH_H_elements[aarx][ctx][rtx], // sic
...@@ -1723,7 +1722,7 @@ static void nr_dlsch_mmse(uint32_t rx_size_symbol, ...@@ -1723,7 +1722,7 @@ static void nr_dlsch_mmse(uint32_t rx_size_symbol,
// Add noise_var such that: H^h * H + noise_var * I // Add noise_var such that: H^h * H + noise_var * I
if (noise_var != 0) { if (noise_var != 0) {
simde__m128i nvar_128i = simde_mm_set1_epi32(noise_var >> 3); simde__m128i nvar_128i = simde_mm_set1_epi32(noise_var >> 3);
for (int p = 0; p < n_tx; p++) { for (int p = 0; p < nl; p++) {
simde__m128i *conjH_H_128i = (simde__m128i *)conjH_H_elements[0][p][p]; simde__m128i *conjH_H_128i = (simde__m128i *)conjH_H_elements[0][p][p];
for (int k = 0; k < 3 * nb_rb_0; k++) { for (int k = 0; k < 3 * nb_rb_0; k++) {
conjH_H_128i[0] = simde_mm_add_epi32(conjH_H_128i[0], nvar_128i); conjH_H_128i[0] = simde_mm_add_epi32(conjH_H_128i[0], nvar_128i);
...@@ -1734,15 +1733,15 @@ static void nr_dlsch_mmse(uint32_t rx_size_symbol, ...@@ -1734,15 +1733,15 @@ static void nr_dlsch_mmse(uint32_t rx_size_symbol,
//Compute the inverse and determinant of the H^*H matrix //Compute the inverse and determinant of the H^*H matrix
//Allocate the inverse matrix //Allocate the inverse matrix
c16_t *inv_H_h_H[n_tx][n_tx]; c16_t *inv_H_h_H[nl][nl];
c16_t inv_H_h_H_data[n_tx][n_tx][12 * nb_rb_0]; c16_t inv_H_h_H_data[nl][nl][12 * nb_rb_0];
memset(inv_H_h_H_data, 0, sizeof(inv_H_h_H_data)); memset(inv_H_h_H_data, 0, sizeof(inv_H_h_H_data));
for (int rtx = 0; rtx < n_tx; rtx++) for (int rtx = 0; rtx < nl; rtx++)
for (int ctx = 0; ctx < n_tx; ctx++) for (int ctx = 0; ctx < nl; ctx++)
inv_H_h_H[ctx][rtx] = inv_H_h_H_data[ctx][rtx]; inv_H_h_H[ctx][rtx] = inv_H_h_H_data[ctx][rtx];
int fp_flag = 1;//0: float point calc 1: Fixed point calc int fp_flag = 1;//0: float point calc 1: Fixed point calc
nr_matrix_inverse(n_tx, nr_matrix_inverse(nl,
conjH_H_elements[0], // Input matrix conjH_H_elements[0], // Input matrix
inv_H_h_H, // Inverse inv_H_h_H, // Inverse
determ_fin, // determin determ_fin, // determin
...@@ -1753,30 +1752,33 @@ static void nr_dlsch_mmse(uint32_t rx_size_symbol, ...@@ -1753,30 +1752,33 @@ static void nr_dlsch_mmse(uint32_t rx_size_symbol,
// multiply Matrix inversion pf H_h_H by the rx signal vector // multiply Matrix inversion pf H_h_H by the rx signal vector
c16_t outtemp[12 * nb_rb_0] __attribute__((aligned(32))); c16_t outtemp[12 * nb_rb_0] __attribute__((aligned(32)));
//Allocate rxdataF for zforcing out //Allocate rxdataF for zforcing out
c16_t rxdataF_zforcing[n_tx][12 * nb_rb_0]; c16_t rxdataF_zforcing[nl][12 * nb_rb_0];
memset(rxdataF_zforcing, 0, sizeof(rxdataF_zforcing)); memset(rxdataF_zforcing, 0, sizeof(rxdataF_zforcing));
for (int rtx=0;rtx<n_tx;rtx++) {//Output Layers row for (int rtx = 0; rtx < nl; rtx++) {//Output Layers row
// loop over Layers rtx=0,...,N_Layers-1 // loop over Layers rtx=0,...,N_Layers-1
for (int ctx = 0; ctx < n_tx; ctx++) { // column multi for (int ctx = 0; ctx < nl; ctx++) { // column multi
// printf("Computing r_%d c_%d\n",rtx,ctx); // printf("Computing r_%d c_%d\n",rtx,ctx);
// print_shorts(" H_h_H=",(int16_t*)&conjH_H_elements[ctx*n_tx+rtx][0][0]); // print_shorts(" H_h_H=",(int16_t*)&conjH_H_elements[ctx*nl+rtx][0][0]);
// print_shorts(" Inv_H_h_H=",(int16_t*)&inv_H_h_H[ctx*n_tx+rtx][0]); // print_shorts(" Inv_H_h_H=",(int16_t*)&inv_H_h_H[ctx*nl+rtx][0]);
nr_a_mult_b(inv_H_h_H[ctx][rtx], (c16_t *)(rxdataF_comp[ctx][0] + symbol * nb_rb * 12), outtemp, nb_rb_0, shift - (fp_flag == 1 ? 2 : 0)); nr_a_mult_b(inv_H_h_H[ctx][rtx],
nr_a_sum_b(rxdataF_zforcing[rtx], outtemp, (c16_t *)(rxdataF_comp[ctx][0] + symbol * nb_rb * 12),
nb_rb_0); // a =a + b outtemp,
nb_rb_0,
shift - (fp_flag == 1 ? 2 : 0));
nr_a_sum_b(rxdataF_zforcing[rtx], outtemp, nb_rb_0); // a = a + b
} }
#ifdef DEBUG_DLSCH_DEMOD #ifdef DEBUG_DLSCH_DEMOD
printf("Computing layer_%d \n",rtx);; printf("Computing layer_%d \n", rtx);
print_shorts(" Rx signal:=",(int16_t*)&rxdataF_zforcing[rtx][0]); print_shorts(" Rx signal:=", (int16_t*)&rxdataF_zforcing[rtx][0]);
print_shorts(" Rx signal:=",(int16_t*)&rxdataF_zforcing[rtx][4]); print_shorts(" Rx signal:=", (int16_t*)&rxdataF_zforcing[rtx][4]);
print_shorts(" Rx signal:=",(int16_t*)&rxdataF_zforcing[rtx][8]); print_shorts(" Rx signal:=", (int16_t*)&rxdataF_zforcing[rtx][8]);
#endif #endif
} }
//Copy zero_forcing out to output array //Copy zero_forcing out to output array
for (int rtx=0;rtx<n_tx;rtx++) for (int rtx = 0; rtx < nl; rtx++)
nr_element_sign(rxdataF_zforcing[rtx], (c16_t *)(rxdataF_comp[rtx][0] + symbol * nb_rb * 12), nb_rb_0, +1); nr_element_sign(rxdataF_zforcing[rtx], (c16_t *)(rxdataF_comp[rtx][0] + symbol * nb_rb * 12), nb_rb_0, + 1);
//Update LLR thresholds with the Matrix determinant //Update LLR thresholds with the Matrix determinant
simde__m128i *dl_ch_mag128_0=NULL,*dl_ch_mag128b_0=NULL,*dl_ch_mag128r_0=NULL,*determ_fin_128; simde__m128i *dl_ch_mag128_0=NULL,*dl_ch_mag128b_0=NULL,*dl_ch_mag128r_0=NULL,*determ_fin_128;
...@@ -1785,7 +1787,7 @@ static void nr_dlsch_mmse(uint32_t rx_size_symbol, ...@@ -1785,7 +1787,7 @@ static void nr_dlsch_mmse(uint32_t rx_size_symbol,
short nr_realpart[8]__attribute__((aligned(16))) = {1,0,1,0,1,0,1,0}; short nr_realpart[8]__attribute__((aligned(16))) = {1,0,1,0,1,0,1,0};
determ_fin_128 = (simde__m128i *)&determ_fin[0]; determ_fin_128 = (simde__m128i *)&determ_fin[0];
if (mod_order>2) { if (mod_order > 2) {
if (mod_order == 4) { if (mod_order == 4) {
QAM_amp128 = simde_mm_set1_epi16(QAM16_n1); //2/sqrt(10) QAM_amp128 = simde_mm_set1_epi16(QAM16_n1); //2/sqrt(10)
QAM_amp128b = simde_mm_setzero_si128(); QAM_amp128b = simde_mm_setzero_si128();
...@@ -1803,25 +1805,24 @@ static void nr_dlsch_mmse(uint32_t rx_size_symbol, ...@@ -1803,25 +1805,24 @@ static void nr_dlsch_mmse(uint32_t rx_size_symbol,
dl_ch_mag128b_0 = (simde__m128i *)dl_ch_magb[0][0]; dl_ch_mag128b_0 = (simde__m128i *)dl_ch_magb[0][0];
dl_ch_mag128r_0 = (simde__m128i *)dl_ch_magr[0][0]; dl_ch_mag128r_0 = (simde__m128i *)dl_ch_magr[0][0];
for (int rb=0; rb<3*nb_rb_0; rb++) { for (int rb = 0; rb < 3 * nb_rb_0; rb++) {
//for symmetric H_h_H matrix, the determinant is only real values //for symmetric H_h_H matrix, the determinant is only real values
mmtmpD2 = simde_mm_sign_epi16(determ_fin_128[0],*(simde__m128i*)&nr_realpart[0]);//set imag part to 0 mmtmpD2 = simde_mm_sign_epi16(determ_fin_128[0],*(simde__m128i*)&nr_realpart[0]);//set imag part to 0
mmtmpD3 = simde_mm_shufflelo_epi16(mmtmpD2,SIMDE_MM_SHUFFLE(2,3,0,1)); mmtmpD3 = simde_mm_shufflelo_epi16(mmtmpD2,SIMDE_MM_SHUFFLE(2,3,0,1));
mmtmpD3 = simde_mm_shufflehi_epi16(mmtmpD3,SIMDE_MM_SHUFFLE(2,3,0,1)); mmtmpD3 = simde_mm_shufflehi_epi16(mmtmpD3,SIMDE_MM_SHUFFLE(2,3,0,1));
mmtmpD2 = simde_mm_add_epi16(mmtmpD2,mmtmpD3); mmtmpD2 = simde_mm_add_epi16(mmtmpD2,mmtmpD3);
dl_ch_mag128_0[0] = mmtmpD2; dl_ch_mag128_0[0] = mmtmpD2;
dl_ch_mag128b_0[0] = mmtmpD2; dl_ch_mag128b_0[0] = mmtmpD2;
dl_ch_mag128r_0[0] = mmtmpD2; dl_ch_mag128r_0[0] = mmtmpD2;
dl_ch_mag128_0[0] = simde_mm_mulhi_epi16(dl_ch_mag128_0[0],QAM_amp128); dl_ch_mag128_0[0] = simde_mm_mulhi_epi16(dl_ch_mag128_0[0],QAM_amp128);
dl_ch_mag128_0[0] = simde_mm_slli_epi16(dl_ch_mag128_0[0],1); dl_ch_mag128_0[0] = simde_mm_slli_epi16(dl_ch_mag128_0[0],1);
dl_ch_mag128b_0[0] = simde_mm_mulhi_epi16(dl_ch_mag128b_0[0],QAM_amp128b); dl_ch_mag128b_0[0] = simde_mm_mulhi_epi16(dl_ch_mag128b_0[0],QAM_amp128b);
dl_ch_mag128b_0[0] = simde_mm_slli_epi16(dl_ch_mag128b_0[0],1); dl_ch_mag128b_0[0] = simde_mm_slli_epi16(dl_ch_mag128b_0[0],1);
dl_ch_mag128r_0[0] = simde_mm_mulhi_epi16(dl_ch_mag128r_0[0],QAM_amp128r); dl_ch_mag128r_0[0] = simde_mm_mulhi_epi16(dl_ch_mag128r_0[0],QAM_amp128r);
dl_ch_mag128r_0[0] = simde_mm_slli_epi16(dl_ch_mag128r_0[0],1); dl_ch_mag128r_0[0] = simde_mm_slli_epi16(dl_ch_mag128r_0[0],1);
determ_fin_128 += 1; determ_fin_128 += 1;
dl_ch_mag128_0 += 1; dl_ch_mag128_0 += 1;
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment