Improved polar encoder

openair1/PHY/CODING/nrPolar_tools/nr_polar_defs.h

@@ -284,6 +284,8 @@ void nr_polar_generate_u(uint64_t *u,
 
 void nr_polar_uxG(uint64_t *D, const uint64_t *u, const fourDimArray_t *G_N_tab, uint16_t N);
 
+void nr_polar_uxG2(uint8_t* D, uint8_t const* u, size_t N);
+
 void nr_polar_info_extraction_from_u(uint64_t *Cprime,
                                      const uint8_t *u,
                                      const uint8_t *information_bit_pattern,

openair1/PHY/CODING/nrPolar_tools/nr_polar_encoder.c

@@ -485,12 +485,13 @@ void build_polar_tables(t_nrPolar_params *polarParams) {
     polarParams->rm_tab[tcnt] = polarParams->rate_matching_pattern[outpos] >> shift;
 }
 
+
 void polar_encoder_fast(uint64_t *A,
                         void *out,
-                        int32_t crcmask,
-                        uint8_t ones_flag,
-                        int8_t messageType,
-                        uint16_t messageLength,
+                        int32_t crcmask, 
+                        uint8_t ones_flag,  
+                        int8_t messageType, 
+                        uint16_t messageLength, 
                         uint8_t aggregation_level) {
                         
   t_nrPolar_params *polarParams=nr_polar_params(messageType, messageLength, aggregation_level, false);
@@ -687,8 +688,9 @@ void polar_encoder_fast(uint64_t *A,
   printf("\n");
 #endif
 
-  uint64_t D[8] = {0};
-  nr_polar_uxG(D, u, polarParams->G_N_tab, polarParams->N);
+  uint64_t D[8];
+//  nr_polar_uxG(D, u, polarParams->G_N_tab, polarParams->N);
+  nr_polar_uxG2((uint8_t*)D, (uint8_t*)u, polarParams->N);
 
 #ifdef POLAR_CODING_DEBUG
   printf("d: ");
@@ -701,6 +703,7 @@ void polar_encoder_fast(uint64_t *A,
     printf("%lu", (D[n1] >> n2) & 1);
   }
   printf("\n");
+  fflush(stdout);
 #endif
 
   memset((void*)out,0,polarParams->encoderLength>>3);

openair1/PHY/CODING/nrPolar_tools/nr_polar_procedures.c

@@ -32,6 +32,8 @@
 
 #include "PHY/CODING/nrPolar_tools/nr_polar_defs.h"
 
+#include <stdlib.h>
+#include <stdint.h>
 
 // TS 38.212 - Section 5.3.1.2 Polar encoding
 void nr_polar_generate_u(uint64_t *u,
@@ -119,6 +121,107 @@ void nr_polar_info_extraction_from_u(uint64_t *Cprime,
   }
 }
 
+
+static
+void encode_packed_byte(uint8_t* in_out)
+{
+  //    *in_out ^= 0xAA & (*in_out << 1);
+  //    *in_out ^= 0xCC & (*in_out << 2);
+  //    *in_out ^= *in_out  in_out << 4;
+  //
+  *in_out ^= 0x55 & (*in_out >> 1);
+  *in_out ^= 0x33 & (*in_out >> 2);
+  *in_out ^= *in_out >> 4;
+}
+
+static
+void polar_encode_bits(uint8_t* in_out, size_t N)
+{
+  int num_bytes_per_block = N >> 3;
+  for(int i = 0; i < num_bytes_per_block; ++i){
+    encode_packed_byte(&in_out[i]);
+  }
+}
+
+static 
+uint32_t log2_floor(uint32_t x) 
+{
+  if(x == 0)
+    return 0; 
+  uint32_t clz = __builtin_clz(x);
+  return 31U - clz;
+}
+
+static
+void polar_encode_bytes(uint8_t* in_out, size_t N)
+{
+  int brnch_sz = 1;
+  int n_brnch = N >> 4;
+  size_t const blck_pwr = log2_floor(N); 
+  for (int stage = 3; stage < blck_pwr; ++stage) {
+    for (int brnch = 0; brnch < n_brnch; ++brnch) {
+        for(int byte = 0; byte < brnch_sz; ++byte){
+          size_t const dst = 2*brnch_sz*brnch + byte;
+          in_out[dst] ^= in_out[dst + brnch_sz];
+        }
+    }
+    n_brnch >>= 1;
+    brnch_sz <<= 1;
+  }
+}
+
+static const 
+uint8_t reverse_tbl[] = {
+        0x00, 0x80, 0x40, 0xc0, 0x20, 0xa0, 0x60, 0xe0,
+        0x10, 0x90, 0x50, 0xd0, 0x30, 0xb0, 0x70, 0xf0,
+        0x08, 0x88, 0x48, 0xc8, 0x28, 0xa8, 0x68, 0xe8,
+        0x18, 0x98, 0x58, 0xd8, 0x38, 0xb8, 0x78, 0xf8,
+        0x04, 0x84, 0x44, 0xc4, 0x24, 0xa4, 0x64, 0xe4,
+        0x14, 0x94, 0x54, 0xd4, 0x34, 0xb4, 0x74, 0xf4,
+        0x0c, 0x8c, 0x4c, 0xcc, 0x2c, 0xac, 0x6c, 0xec,
+        0x1c, 0x9c, 0x5c, 0xdc, 0x3c, 0xbc, 0x7c, 0xfc,
+        0x02, 0x82, 0x42, 0xc2, 0x22, 0xa2, 0x62, 0xe2,
+        0x12, 0x92, 0x52, 0xd2, 0x32, 0xb2, 0x72, 0xf2,
+        0x0a, 0x8a, 0x4a, 0xca, 0x2a, 0xaa, 0x6a, 0xea,
+        0x1a, 0x9a, 0x5a, 0xda, 0x3a, 0xba, 0x7a, 0xfa,
+        0x06, 0x86, 0x46, 0xc6, 0x26, 0xa6, 0x66, 0xe6,
+        0x16, 0x96, 0x56, 0xd6, 0x36, 0xb6, 0x76, 0xf6,
+        0x0e, 0x8e, 0x4e, 0xce, 0x2e, 0xae, 0x6e, 0xee,
+        0x1e, 0x9e, 0x5e, 0xde, 0x3e, 0xbe, 0x7e, 0xfe,
+        0x01, 0x81, 0x41, 0xc1, 0x21, 0xa1, 0x61, 0xe1,
+        0x11, 0x91, 0x51, 0xd1, 0x31, 0xb1, 0x71, 0xf1,
+        0x09, 0x89, 0x49, 0xc9, 0x29, 0xa9, 0x69, 0xe9,
+        0x19, 0x99, 0x59, 0xd9, 0x39, 0xb9, 0x79, 0xf9,
+        0x05, 0x85, 0x45, 0xc5, 0x25, 0xa5, 0x65, 0xe5,
+        0x15, 0x95, 0x55, 0xd5, 0x35, 0xb5, 0x75, 0xf5,
+        0x0d, 0x8d, 0x4d, 0xcd, 0x2d, 0xad, 0x6d, 0xed,
+        0x1d, 0x9d, 0x5d, 0xdd, 0x3d, 0xbd, 0x7d, 0xfd,
+        0x03, 0x83, 0x43, 0xc3, 0x23, 0xa3, 0x63, 0xe3,
+        0x13, 0x93, 0x53, 0xd3, 0x33, 0xb3, 0x73, 0xf3,
+        0x0b, 0x8b, 0x4b, 0xcb, 0x2b, 0xab, 0x6b, 0xeb,
+        0x1b, 0x9b, 0x5b, 0xdb, 0x3b, 0xbb, 0x7b, 0xfb,
+        0x07, 0x87, 0x47, 0xc7, 0x27, 0xa7, 0x67, 0xe7,
+        0x17, 0x97, 0x57, 0xd7, 0x37, 0xb7, 0x77, 0xf7,
+        0x0f, 0x8f, 0x4f, 0xcf, 0x2f, 0xaf, 0x6f, 0xef,
+        0x1f, 0x9f, 0x5f, 0xdf, 0x3f, 0xbf, 0x7f, 0xff,
+    };
+
+void nr_polar_uxG2(uint8_t* D2, uint8_t const* u, size_t N)
+{
+  uint8_t tmp[N/8];
+  for(int i = 0; i < N/8; ++i)
+    tmp[i] = u[N/8 - 1 - i];
+
+  for(int i = 0; i < N/8; ++i)
+    D2[i] = reverse_tbl[tmp[i]];
+
+  // Do the encoding/xor for the bottom 3 levels of the tree.
+  // Thus, data remaining to encode, is in 2^3 type.
+  polar_encode_bits(D2, N);
+  // Xor the remaining tree levels. Use bytes for it
+  polar_encode_bytes(D2, N);
+}
+
 void nr_polar_uxG(uint64_t *D, const uint64_t *u, const fourDimArray_t *G_N_tab, uint16_t N)
 {
   const int N64 = N / 64;