Merge branch 'l1-sidelink' of https://gitlab.eurecom.fr/matzakos/LTE-D2D into l1-sidelink

openair1/PHY/LTE_ESTIMATION/lte_sync_time.c

@@ -43,8 +43,6 @@
 int32_t* sync_corr_ue0 = NULL;
 int32_t* sync_corr_ue1 = NULL;
 int32_t* sync_corr_ue2 = NULL;
-int32_t sync_tmp[2048*4] __attribute__((aligned(32)));
-int16_t syncF_tmp[2048*2] __attribute__((aligned(32)));
 
 
 extern int16_t s6n_kHz_7_5[1920];
@@ -63,6 +61,8 @@ int lte_sync_time_init(LTE_DL_FRAME_PARMS *frame_parms )   // LTE_UE_COMMON *com
 {
 
   int i,k;
+  int32_t sync_tmp[2048*4] __attribute__((aligned(32)));
+  int16_t syncF_tmp[2048*2] __attribute__((aligned(32)));
 
   sync_corr_ue0 = (int32_t *)malloc16(4*LTE_NUMBER_OF_SUBFRAMES_PER_FRAME*sizeof(int32_t)*frame_parms->samples_per_tti);
   sync_corr_ue1 = (int32_t *)malloc16(4*LTE_NUMBER_OF_SUBFRAMES_PER_FRAME*sizeof(int32_t)*frame_parms->samples_per_tti);
@@ -167,7 +167,7 @@ int lte_sync_time_init(LTE_DL_FRAME_PARMS *frame_parms )   // LTE_UE_COMMON *com
   } else  AssertFatal(1==0,"primary_synch1SL_time_rx not allocated\n");
 
 
-
+  memset((void*)syncF_tmp,0,2048*sizeof(int32_t));
   // generate oversampled sync_time sequences
   k=frame_parms->ofdm_symbol_size-36;
 
@@ -217,6 +217,8 @@ int lte_sync_time_init(LTE_DL_FRAME_PARMS *frame_parms )   // LTE_UE_COMMON *com
   for (i=0; i<frame_parms->ofdm_symbol_size; i++)
     ((int32_t*)primary_synch0_time)[i] = sync_tmp[i];
 
+  memset((void*)syncF_tmp,0,2048*sizeof(int32_t));
+
   k=frame_parms->ofdm_symbol_size-36;
 
   for (i=0; i<72; i++) {
@@ -265,6 +267,8 @@ int lte_sync_time_init(LTE_DL_FRAME_PARMS *frame_parms )   // LTE_UE_COMMON *com
   for (i=0; i<frame_parms->ofdm_symbol_size; i++)
     ((int32_t*)primary_synch1_time)[i] = sync_tmp[i];
 
+  memset((void*)syncF_tmp,0,2048*sizeof(int32_t));
+
   k=frame_parms->ofdm_symbol_size-36;
 
   for (i=0; i<72; i++) {
@@ -313,6 +317,8 @@ int lte_sync_time_init(LTE_DL_FRAME_PARMS *frame_parms )   // LTE_UE_COMMON *com
   for (i=0; i<frame_parms->ofdm_symbol_size; i++)
     ((int32_t*)primary_synch2_time)[i] = sync_tmp[i];
 
+  memset((void*)syncF_tmp,0,2048*sizeof(int32_t));
+
   k=frame_parms->ofdm_symbol_size-36;
 
   for (i=0; i<72; i++) {
@@ -325,7 +331,7 @@ int lte_sync_time_init(LTE_DL_FRAME_PARMS *frame_parms )   // LTE_UE_COMMON *com
     }
   }
 
-  uint16_t *kHz7_5ptr;
+  int16_t *kHz7_5ptr;
 
 
   switch (frame_parms->N_RB_DL) {
@@ -333,21 +339,21 @@ int lte_sync_time_init(LTE_DL_FRAME_PARMS *frame_parms )   // LTE_UE_COMMON *com
     idft128((int16_t*)syncF_tmp,          /// complex input
 	   (int16_t*)sync_tmp, /// complex output
 	   1);
-    kHz7_5ptr = (frame_parms->Ncp==0) ? (((uint32_t*)s6n_kHz_7_5)+138): (((uint32_t*)s6e_kHz_7_5)+160);
+    kHz7_5ptr = (frame_parms->Ncp==0) ? &s6n_kHz_7_5[2*138]: &s6e_kHz_7_5[2*160];
 
     break;
   case 25:
     idft512((int16_t*)syncF_tmp,          /// complex input
 	   (int16_t*)sync_tmp, /// complex output
 	   1);
-    kHz7_5ptr = (frame_parms->Ncp==0) ? (((uint32_t*)s25n_kHz_7_5)+552) : (((uint32_t*)s25e_kHz_7_5)+640);
+    kHz7_5ptr = (frame_parms->Ncp==0) ? &s25n_kHz_7_5[2*552] : &s25e_kHz_7_5[2*640];
 
     break;
   case 50:
     idft1024((int16_t*)syncF_tmp,          /// complex input
 	    (int16_t*)sync_tmp, /// complex output
 	    1);
-    kHz7_5ptr = (frame_parms->Ncp==0) ? &s50n_kHz_7_5[2*1104] : s50e_kHz_7_5[2*1280];
+    kHz7_5ptr = (frame_parms->Ncp==0) ? &s50n_kHz_7_5[2*1104] : &s50e_kHz_7_5[2*1280];
     printf("%p\n",kHz7_5ptr);
     break;
     
@@ -355,14 +361,14 @@ int lte_sync_time_init(LTE_DL_FRAME_PARMS *frame_parms )   // LTE_UE_COMMON *com
     idft1536((int16_t*)syncF_tmp,          /// complex input
 	     (int16_t*)sync_tmp,
 	     1); /// complex output
-    kHz7_5ptr = (frame_parms->Ncp==0) ? (((uint32_t*)s75n_kHz_7_5)+1656) : (((uint32_t*)s75e_kHz_7_5)+1920);
+    kHz7_5ptr = (frame_parms->Ncp==0) ? &s75n_kHz_7_5[2*1656]: &s75e_kHz_7_5[2*1920];
 
     break;
   case 100:
     idft2048((int16_t*)syncF_tmp,          /// complex input
 	     (int16_t*)sync_tmp, /// complex output
 	     1);
-    kHz7_5ptr = (frame_parms->Ncp==0) ? (((uint32_t*)s100n_kHz_7_5)+2208) : (((uint32_t*)s100e_kHz_7_5)+2560);
+    kHz7_5ptr = (frame_parms->Ncp==0) ? &s100n_kHz_7_5[2*2208] : &s100e_kHz_7_5[2*2560];
 
     break;
   default:
@@ -380,6 +386,7 @@ int lte_sync_time_init(LTE_DL_FRAME_PARMS *frame_parms )   // LTE_UE_COMMON *com
     primary_synch0SL_time_rx[1+(i<<1)] = (int16_t)(((int32_t)primary_synch0SL_time[imod<<1]*kHz7_5ptr[1+(i<<1)])>>15) + (int16_t)(((int32_t)primary_synch0SL_time[1+(imod<<1)]*kHz7_5ptr[i<<1])>>15);
   }
  
+  memset((void*)syncF_tmp,0,2048*sizeof(int32_t));
 
   k=frame_parms->ofdm_symbol_size-36;
 
@@ -425,28 +432,40 @@ int lte_sync_time_init(LTE_DL_FRAME_PARMS *frame_parms )   // LTE_UE_COMMON *com
     break;
   }
 
+  for (i=0;i<(frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples);i++)
+      ((int32_t*)primary_synch1SL_time)[i] = sync_tmp[(i+(frame_parms->ofdm_symbol_size-frame_parms->nb_prefix_samples))%frame_parms->ofdm_symbol_size];
+
   for (i=0; i<(frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples)*2; i++) {
     imod = i%(frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples);
-    if (i<(frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples))
-      ((int32_t*)primary_synch1SL_time)[i] = sync_tmp[(i+(frame_parms->ofdm_symbol_size-frame_parms->nb_prefix_samples))%frame_parms->ofdm_symbol_size];
 
-    primary_synch1SL_time_rx[i<<1]     = (int16_t)(((int32_t)primary_synch1SL_time[imod<<1]*kHz7_5ptr[i<<1])>>15) - 
+    primary_synch1SL_time_rx[i<<1]     = (int16_t)(((int32_t)primary_synch1SL_time[imod<<1]*kHz7_5ptr[i<<1])>>15) + 
                                          (int16_t)(((int32_t)primary_synch1SL_time[1+(imod<<1)]*kHz7_5ptr[1+(i<<1)])>>15);
-    primary_synch1SL_time_rx[1+(i<<1)] = (int16_t)(((int32_t)primary_synch1SL_time[imod<<1]*kHz7_5ptr[1+(i<<1)])>>15) +
+    primary_synch1SL_time_rx[1+(i<<1)] = -(int16_t)(((int32_t)primary_synch1SL_time[imod<<1]*kHz7_5ptr[1+(i<<1)])>>15) +
                                          (int16_t)(((int32_t)primary_synch1SL_time[1+(imod<<1)]*kHz7_5ptr[i<<1])>>15);
+ /*   printf("sync_timeSL1(%d) : (%d,%d) x (%d,%d)' = (%d,%d)\n",
+          i,
+          primary_synch1SL_time[imod<<1],
+          primary_synch1SL_time[1+(imod<<1)],
+          kHz7_5ptr[i<<1],
+          kHz7_5ptr[1+(i<<1)],
+          primary_synch1SL_time_rx[i<<1],
+          primary_synch1SL_time_rx[1+(i<<1)]);
+*/
   }
 
 
   /*
   write_output("primary_sync0.m","psync0",primary_synch0_time,frame_parms->ofdm_symbol_size,1,1);
   write_output("primary_sync1.m","psync1",primary_synch1_time,frame_parms->ofdm_symbol_size,1,1);
-  write_output("primary_sync2.m","psync2",primary_synch2_time,frame_parms->ofdm_symbol_size,1,1);
-  write_output("primary_syncSL0.m","psyncSL0",primary_synch0SL_time,frame_parms->ofdm_symbol_size,1,1);
+  write_output("primary_sync2.m","psync2",primary_synch2_time,frame_parms->ofdm_symbol_size,1,1);*/
+  write_output("primary_syncSL0.m","psyncSL0",primary_synch0SL_time,frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples,1,1);
+  write_output("primary_syncSL1.m","psyncSL1",primary_synch1SL_time,frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples,1,1);
 
-  write_output("primary_syncSL1.m","psyncSL1",primary_synch1SL_time_rx,2*(frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples),1,1);
-  write_output("primary_syncSL0.m","psyncSL0",primary_synch0SL_time_rx,2*(frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples),1,1);
+    
+  write_output("primary_syncSL1rx.m","psyncSL1rx",primary_synch1SL_time_rx,2*(frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples),1,1);
+  write_output("primary_syncSL0rx.m","psyncSL0rx",primary_synch0SL_time_rx,2*(frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples),1,1);
   write_output("kHz75.m","kHz75",kHz7_5ptr,2*1096,1,1);
- */ 
+  
   return (1);
 }
 
@@ -509,6 +528,11 @@ static inline int32_t abs32(int32_t x)
   return (((int32_t)((int16_t*)&x)[0])*((int32_t)((int16_t*)&x)[0]) + ((int32_t)((int16_t*)&x)[1])*((int32_t)((int16_t*)&x)[1]));
 }
 
+static inline int64_t abs64(int64_t x)
+{
+  return (((int64_t)((int32_t*)&x)[0])*((int64_t)((int32_t*)&x)[0]) + ((int64_t)((int32_t*)&x)[1])*((int64_t)((int32_t*)&x)[1]));
+}
+
 #ifdef DEBUG_PHY
 int debug_cnt=0;
 #endif
@@ -680,12 +704,12 @@ int lte_sync_timeSL(PHY_VARS_UE *ue,
 							    (void*)&ue->common_vars.rxdata_syncSL[ar][0],
 							    frame_parms->ofdm_symbol_size);
 
-  int32_t tmp0,tmp1;
-  int32_t magtmp0,magtmp1,maxlev0=0,maxlev1=0;
+  int64_t tmp0,tmp1;
+  int64_t magtmp0,magtmp1,maxlev0=0,maxlev1=0;
   int     maxpos0=0,maxpos1=0;
   int64_t avg0=0,avg1=0;
-  int32_t result;
-  int32_t **rxdata = ue->common_vars.rxdata_syncSL; ///rx data in time domain
+  int64_t result;
+  int32_t **rxdata = (int32_t**)ue->common_vars.rxdata_syncSL; ///rx data in time domain
   RU_t ru_tmp;
   int16_t **rxdata_7_5kHz    = ue->sl_rxdata_7_5kHz;
 
@@ -697,39 +721,51 @@ int lte_sync_timeSL(PHY_VARS_UE *ue,
   ru_tmp.common.rxdata_7_5kHz = (int32_t**)rxdata_7_5kHz;
   ru_tmp.nb_rx = frame_parms->nb_antennas_rx;
   
-
-
-  for (int n=0; n<length; n+=4) {
+  int maxval=0;
+  for (int i=0;i<2*(frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples);i++) {
+    maxval = max(maxval,primary_synch0SL_time_rx[i]);
+    maxval = max(maxval,-primary_synch0SL_time_rx[i]);
+    maxval = max(maxval,primary_synch1SL_time_rx[i]);
+    maxval = max(maxval,-primary_synch1SL_time_rx[i]);
+  }
+  int shift = log2_approx(maxval);//*(frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples)*2);
+  printf("Synchtime SL : shifting by %d bits\n",shift);
+  for (int n=0; n<length; n+=4) 
+  {
 
     tmp0 = 0;
     tmp1 = 0;
+    int32_t tmp0_re=((int32_t*)&tmp0)[0], tmp0_im=((int32_t*)&tmp0)[1];
+    int32_t tmp1_re=((int32_t*)&tmp1)[0], tmp1_im=((int32_t*)&tmp1)[1];
 
     //calculate dot product of primary_synch0_time and rxdata[ar][n] (ar=0..nb_ant_rx) and store the sum in temp[n];
     for (int ar=0; ar<frame_parms->nb_antennas_rx; ar++) {
       
-      result  = dot_product((int16_t*)primary_synch0SL_time_rx,
+      result  = dot_product(primary_synch0SL_time_rx,
 			    (int16_t*) &(rxdata[ar][n]),
 			    (frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples)*2,
-			    11);
+			    shift);
       
-      ((int16_t*)&tmp0)[0] += ((int16_t*) &result)[0];
-      ((int16_t*)&tmp0)[1] += ((int16_t*) &result)[1];
+      tmp0_re += ((int32_t*) &result)[0];
+      tmp0_im += ((int32_t*) &result)[1];
 
-      result  = dot_product((int16_t*)primary_synch1SL_time_rx,
+      result  = dot_product(primary_synch1SL_time_rx,
 			    (int16_t*) &(rxdata[ar][n]),
 			    (frame_parms->ofdm_symbol_size+frame_parms->nb_prefix_samples)*2,
-			    11);
+			    shift);
 
-      ((int16_t*)&tmp1)[0] += ((int16_t*) &result)[0];
-      ((int16_t*)&tmp1)[1] += ((int16_t*) &result)[1];
+      tmp1_re += ((int32_t*) &result)[0];
+      tmp1_im += ((int32_t*) &result)[1];
 
     }
 
     // tmpi holds <synchi,rx0>+<synci,rx1>+...+<synchi,rx_{nbrx-1}>
 
-    magtmp0 = abs32(tmp0);
-    magtmp1 = abs32(tmp1);
+    magtmp0 = (int64_t)tmp0_re*tmp0_re + (int64_t)tmp0_im*tmp0_im;
+    magtmp1 = (int64_t)tmp1_re*tmp1_re + (int64_t)tmp1_im*tmp1_im;
 
+    //printf("0: n %d (%d,%d) => %lld\n",n,tmp0_re,tmp0_im,magtmp0);
+    //printf("1: n %d (%d,%d) => %lld\n",n,tmp1_re,tmp1_im,magtmp1);
     // this does max |tmpi(n)|^2 + |tmpi(n-L)|^2 and argmax |tmpi(n)|^2 + |tmpi(n-L)|^2
     
     if (magtmp0>maxlev0) { maxlev0 = magtmp0; maxpos0 = n; }
@@ -737,8 +773,8 @@ int lte_sync_timeSL(PHY_VARS_UE *ue,
     avg0 += magtmp0;
     avg1 += magtmp1;
     if (n<4*FRAME_LENGTH_COMPLEX_SAMPLES) {
-      sync_corr_ue1[n] = magtmp1;       
-      sync_corr_ue0[n] = magtmp0;       
+      sync_corr_ue1[n] = (int32_t)(magtmp1>>31);       
+      sync_corr_ue0[n] = (int32_t)(magtmp0>>31);       
     }
   }
   avg0/=(length/4);
@@ -746,7 +782,8 @@ int lte_sync_timeSL(PHY_VARS_UE *ue,
 
   // PSS in symbol 1
   int pssoffset = frame_parms->ofdm_symbol_size + frame_parms->nb_prefix_samples0;
-  printf("maxpos1 = %d, pssoffset = %d\n",maxpos1,pssoffset);
+  printf("maxpos = (%d,%d), pssoffset = %d, maxlev= (%lld,%lld) avglev (%lld,%lld)\n",maxpos0,maxpos1,pssoffset,
+         (long long int)maxlev0,(long long int)maxlev1,(long long int)avg0,(long long int)avg1);
  
   if (maxlev0 > maxlev1) {
     if ((int64_t)maxlev0 > (5*avg0)) {*lev = maxlev0; *ind=0; *avg=avg0; return((length+maxpos0-pssoffset)%length);};

openair1/PHY/LTE_TRANSPORT/initial_syncSL.c

@@ -50,8 +50,8 @@ int initial_syncSL(PHY_VARS_UE *ue) {
 				    &index,
 				    &psslevel,
 				    &avglevel);
-  printf("index %d, psslevel %lld dB avglevel %lld dB => %d sample offset\n",
-	 index,dB_fixed(psslevel),dB_fixed(avglevel),ue->rx_offsetSL);
+  printf("index %d, psslevel %d dB avglevel %d dB => %d sample offset\n",
+	 index,dB_fixed64((uint64_t)psslevel),dB_fixed64((uint64_t)avglevel),ue->rx_offsetSL);
   if (ue->rx_offsetSL >= 0) {
     int32_t sss_metric;
     int32_t phase_max;
@@ -60,13 +60,11 @@ int initial_syncSL(PHY_VARS_UE *ue) {
   
     if (rx_psbch(ue) == -1) {
       ue->slbch_errors++;
-/*
       write_output("rxsig0.m","rxs0",&ue->common_vars.rxdata_syncSL[0][0],40*ue->frame_parms.samples_per_tti,1,1);
       write_output("corr0.m","rxsync0",sync_corr_ue0,40*ue->frame_parms.samples_per_tti,1,2);
       write_output("corr1.m","rxsync1",sync_corr_ue1,40*ue->frame_parms.samples_per_tti,1,2);
 
       exit(-1);
-*/
       return(-1);
     }
     else {

openair1/PHY/LTE_TRANSPORT/slsss.c

@@ -359,10 +359,10 @@ int rx_slsss(PHY_VARS_UE *ue,int32_t *tot_metric,uint8_t *phase_max,int Nid2)
 		    0);
 
   //  write_output("rxdataF0.m","rxF0",&rxdataF[0][0],2*14*ue->frame_parms.ofdm_symbol_size,1,1);
-  /*  write_output("pss0_ext.m","pss0ext",pss0_ext,72,1,1);
+    write_output("pss0_ext.m","pss0ext",pss0_ext,72,1,1);
     write_output("sss0_ext.m","sss0ext",sss0_ext,72,1,1);
     write_output("pss1_ext.m","pss1ext",pss1_ext,72,1,1);
-    write_output("sss1_ext.m","sss1ext",sss1_ext,72,1,1); */
+    write_output("sss1_ext.m","sss1ext",sss1_ext,72,1,1); 
 
 
 
@@ -400,9 +400,10 @@ int rx_slsss(PHY_VARS_UE *ue,int32_t *tot_metric,uint8_t *phase_max,int Nid2)
       sss0_comp16[i] = (int16_t)(sss0comp[i]>>shift);
       sss1_comp16[i] = (int16_t)(sss1comp[i]>>shift);
   }
-    /*write_output("sss0_comp0.m","sss0comp0",sss0_comp16,72,1,1);
-    write_output("sss1_comp0.m","sss1comp0",sss1_comp16,72,1,1);*/
-  //  exit(-1);
+/*
+    write_output("sss0_comp0.m","sss0comp0",sss0_comp16,72,1,1);
+    write_output("sss1_comp0.m","sss1comp0",sss1_comp16,72,1,1);
+    exit(-1); */
   // now do the SSS detection based on the precomputed sequences in PHY/LTE_TRANSPORT/sss.h
 
   *tot_metric = -99999999;

openair1/PHY/TOOLS/cdot_prod.c

@@ -21,7 +21,7 @@
 
 #include "defs.h"
 #include "PHY/sse_intrin.h"
-
+#include <stdio.h>
 // returns the complex dot product of x and y
 
 #ifdef MAIN
@@ -30,7 +30,7 @@ void print_shorts(char *s,__m128i *x);
 void print_bytes(char *s,__m128i *x);
 #endif
 
-int32_t dot_product(int16_t *x,
+int64_t dot_product(int16_t *x,
                     int16_t *y,
                     uint32_t N, //must be a multiple of 8
                     uint8_t output_shift)
@@ -40,7 +40,6 @@ int32_t dot_product(int16_t *x,
 
 #if defined(__x86_64__) || defined(__i386__)
   __m128i *x128,*y128,mmtmp1,mmtmp2,mmtmp3,mmcumul,mmcumul_re,mmcumul_im;
-  __m64 mmtmp7;
   __m128i minus_i = _mm_set_epi16(-1,1,-1,1,-1,1,-1,1);
   int32_t result;
 
@@ -52,37 +51,37 @@ int32_t dot_product(int16_t *x,
 
   for (n=0; n<(N>>2); n++) {
 
-    //printf("n=%d, x128=%p, y128=%p\n",n,x128,y128);
+//    printf("n=%d, x128=%p, y128=%p\n",n,x128,y128);
   //      print_shorts("x",&x128[0]);
     //    print_shorts("y",&y128[0]);
 
     // this computes Re(z) = Re(x)*Re(y) + Im(x)*Im(y)
     mmtmp1 = _mm_madd_epi16(x128[0],y128[0]);
-    //    print_ints("re",&mmtmp1);
+      //  print_ints("retmp",&mmtmp1);
     // mmtmp1 contains real part of 4 consecutive outputs (32-bit)
 
     // shift and accumulate results
     mmtmp1 = _mm_srai_epi32(mmtmp1,output_shift);
     mmcumul_re = _mm_add_epi32(mmcumul_re,mmtmp1);
-    //    print_ints("re",&mmcumul_re);
+        //print_ints("re",&mmcumul_re);
 
 
     // this computes Im(z) = Re(x)*Im(y) - Re(y)*Im(x)
     mmtmp2 = _mm_shufflelo_epi16(y128[0],_MM_SHUFFLE(2,3,0,1));
-    //    print_shorts("y",&mmtmp2);
+        //print_shorts("y",&mmtmp2);
     mmtmp2 = _mm_shufflehi_epi16(mmtmp2,_MM_SHUFFLE(2,3,0,1));
-    //    print_shorts("y",&mmtmp2);
+        //print_shorts("y",&mmtmp2);
     mmtmp2 = _mm_sign_epi16(mmtmp2,minus_i);
           //  print_shorts("y",&mmtmp2);
 
     mmtmp3 = _mm_madd_epi16(x128[0],mmtmp2);
-    //        print_ints("im",&mmtmp3);
+            //print_ints("imtmp",&mmtmp3);
     // mmtmp3 contains imag part of 4 consecutive outputs (32-bit)
 
     // shift and accumulate results
     mmtmp3 = _mm_srai_epi32(mmtmp3,output_shift);
     mmcumul_im = _mm_add_epi32(mmcumul_im,mmtmp3);
-    //    print_ints("im",&mmcumul_im);
+        //print_ints("im",&mmcumul_im);
 
     x128++;
     y128++;
@@ -90,24 +89,18 @@ int32_t dot_product(int16_t *x,
 
   // this gives Re Re Im Im
   mmcumul = _mm_hadd_epi32(mmcumul_re,mmcumul_im);
-  //  print_ints("cumul1",&mmcumul);
+    //print_ints("cumul1",&mmcumul);
 
   // this gives Re Im Re Im
   mmcumul = _mm_hadd_epi32(mmcumul,mmcumul);
 
-  //  print_ints("cumul2",&mmcumul);
+    //print_ints("cumul2",&mmcumul);
 
 
   //mmcumul = _mm_srai_epi32(mmcumul,output_shift);
   // extract the lower half
-  mmtmp7 = _mm_movepi64_pi64(mmcumul);
-  //  print_ints("mmtmp7",&mmtmp7);
-  // pack the result
-  mmtmp7 = _mm_packs_pi32(mmtmp7,mmtmp7);
-  //  print_shorts("mmtmp7",&mmtmp7);
-  // convert back to integer
-  result = _mm_cvtsi64_si32(mmtmp7);
-  
+  result = _mm_extract_epi64(mmcumul,0);
+  //printf("result: (%d,%d)\n",((int32_t*)&result)[0],((int32_t*)&result)[1]); 
   _mm_empty();
   _m_empty();
  

openair1/PHY/TOOLS/dB_routines.c

@@ -23,7 +23,7 @@
 
 // Approximate 10*log10(x) in fixed point : x = 0...(2^32)-1
 
-int8_t dB_table[256] = {
+uint8_t dB_table[256] = {
   0,
   3,
   5,
@@ -282,7 +282,7 @@ int8_t dB_table[256] = {
   24
 };
 
-int16_t dB_table_times10[256] = {
+uint16_t dB_table_times10[256] = {
   0,
   30,
   47,
@@ -571,9 +571,9 @@ int8_t dB_fixed(int x) {
 }
 */
 
-int16_t dB_fixed_times10(uint32_t x)
+uint16_t dB_fixed_times10(uint32_t x)
 {
-  int16_t dB_power=0;
+  uint8_t dB_power=0;
 
 
   if (x==0) {
@@ -597,10 +597,17 @@ int16_t dB_fixed_times10(uint32_t x)
   return dB_power;
 }
 
-int8_t dB_fixed(uint32_t x)
+uint8_t dB_fixed64(uint64_t x)
 {
 
-  int8_t dB_power=0;
+  if (x<(((uint64_t)1)<<32)) return(dB_fixed((uint32_t)x));
+  else                       return(4*dB_table[255] + dB_fixed(x>>32)); 
+ 
+}
+uint8_t dB_fixed(uint32_t x)
+{
+
+  uint8_t dB_power=0;
 
 
   if (x==0) {
@@ -624,7 +631,7 @@ int8_t dB_fixed(uint32_t x)
   return dB_power;
 }
 
-int8_t dB_fixed2(uint32_t x, uint32_t y)
+uint8_t dB_fixed2(uint32_t x, uint32_t y)
 {
 
   if ((x>0) && (y>0) )

openair1/PHY/TOOLS/defs.h

@@ -347,16 +347,18 @@ Compensate the phase rotation of the RF. WARNING: This function is currently unu
 */
 
 
-int8_t dB_fixed(uint32_t x);
+uint8_t dB_fixed(uint32_t x);
 
-int8_t dB_fixed2(uint32_t x,uint32_t y);
+uint8_t dB_fixed2(uint32_t x,uint32_t y);
 
-int16_t dB_fixed_times10(uint32_t x);
+uint8_t dB_fixed64(uint64_t x);
+
+uint16_t dB_fixed_times10(uint32_t x);
 
 int32_t phy_phase_compensation_top (uint32_t pilot_type, uint32_t initial_pilot,
                                     uint32_t last_pilot, int32_t ignore_prefix);
 
-int32_t dot_product(int16_t *x,
+int64_t dot_product(int16_t *x,
                     int16_t *y,
                     uint32_t N, //must be a multiple of 8
                     uint8_t output_shift);

openair1/PHY/extern.h

@@ -60,11 +60,11 @@ extern short primary_synch1SL[144];
 extern unsigned char primary_synch0_tab[72];
 extern unsigned char primary_synch1_tab[72];
 extern unsigned char primary_synch2_tab[72];
-extern const int16_t *primary_synch0_time; //!< index: [0..ofdm_symbol_size*2[
-extern const int16_t *primary_synch1_time; //!< index: [0..ofdm_symbol_size*2[
-extern const int16_t *primary_synch2_time; //!< index: [0..ofdm_symbol_size*2[
-extern const int16_t *primary_synch0SL_time;
-extern const int16_t *primary_synch1SL_time;
+extern int16_t *primary_synch0_time; //!< index: [0..ofdm_symbol_size*2[
+extern int16_t *primary_synch1_time; //!< index: [0..ofdm_symbol_size*2[
+extern int16_t *primary_synch2_time; //!< index: [0..ofdm_symbol_size*2[
+extern int16_t *primary_synch0SL_time;
+extern int16_t *primary_synch1SL_time;
 extern int16_t *primary_synch0SL_time_rx;
 extern int16_t *primary_synch1SL_time_rx;
 

openair1/PHY/vars.h

@@ -33,13 +33,13 @@ char* namepointer_log2;
 
 #include "PHY/LTE_REFSIG/primary_synch.h"
 #include "PHY/LTE_REFSIG/primary_synch_SL.h"
-const int16_t *primary_synch0_time;
-const int16_t *primary_synch1_time;
-const int16_t *primary_synch2_time;
+int16_t *primary_synch0_time;
+int16_t *primary_synch1_time;
+int16_t *primary_synch2_time;
 
 
-const int16_t *primary_synch0SL_time;
-const int16_t *primary_synch1SL_time;
+int16_t *primary_synch0SL_time;
+int16_t *primary_synch1SL_time;
 int16_t *primary_synch0SL_time_rx;
 int16_t *primary_synch1SL_time_rx;