diff --git a/cmake_targets/CMakeLists.txt b/cmake_targets/CMakeLists.txt
index f6fcabf82e93564a36faae081b9cae9320a51d5c..d07694725d0e438cce7832458d7e08115f552645 100644
--- a/cmake_targets/CMakeLists.txt
+++ b/cmake_targets/CMakeLists.txt
@@ -1495,6 +1495,7 @@ set(PHY_SRC_UE
   ${OPENAIR1_DIR}/PHY/NR_TRANSPORT/nr_ulsch_demodulation.c
   ${OPENAIR1_DIR}/PHY/NR_REFSIG/nr_dmrs_rx.c
   ${OPENAIR1_DIR}/PHY/NR_REFSIG/nr_gold.c
+  ${OPENAIR1_DIR}/PHY/NR_REFSIG/scrambling_luts.c
   ${OPENAIR1_DIR}/PHY/NR_REFSIG/dmrs_nr.c
   ${OPENAIR1_DIR}/PHY/NR_REFSIG/ptrs_nr.c
   ${OPENAIR1_DIR}/PHY/NR_UE_ESTIMATION/filt16a_32.c
diff --git a/openair1/PHY/INIT/nr_init.c b/openair1/PHY/INIT/nr_init.c
index 1c0d5a6acadc6d323f04b99569249f52910f4c0f..46fd9fb5fd8909d06e15ead90166ef7ed6e26d64 100644
--- a/openair1/PHY/INIT/nr_init.c
+++ b/openair1/PHY/INIT/nr_init.c
@@ -96,7 +96,6 @@ int phy_init_nr_gNB(PHY_VARS_gNB *gNB,
 
   while(gNB->configured == 0) usleep(10000);
 
-  init_dfts();
   /*
     LOG_I(PHY,"[gNB %"PRIu8"] Initializing DL_FRAME_PARMS : N_RB_DL %"PRIu8", PHICH Resource %d, PHICH Duration %d nb_antennas_tx:%u nb_antennas_rx:%u PRACH[rootSequenceIndex:%u prach_Config_enabled:%u configIndex:%u highSpeed:%u zeroCorrelationZoneConfig:%u freqOffset:%u]\n",
           gNB->Mod_id,
@@ -113,6 +112,8 @@ int phy_init_nr_gNB(PHY_VARS_gNB *gNB,
   LOG_D(PHY,"[MSC_NEW][FRAME 00000][PHY_gNB][MOD %02"PRIu8"][]\n", gNB->Mod_id);
   crcTableInit();
   init_dfts();
+  init_scrambling_luts();
+  init_pucch2_luts();
   load_nrLDPClib();
   // PBCH DMRS gold sequences generation
   nr_init_pbch_dmrs(gNB);
diff --git a/openair1/PHY/NR_REFSIG/nr_refsig.h b/openair1/PHY/NR_REFSIG/nr_refsig.h
index 5af0e880999dcac8e8dec184be57835197e61fcb..d18f6405d7bf2a5f602ca60399cb10a124004495 100644
--- a/openair1/PHY/NR_REFSIG/nr_refsig.h
+++ b/openair1/PHY/NR_REFSIG/nr_refsig.h
@@ -26,7 +26,7 @@
 
 #include "PHY/defs_gNB.h"
 #include "PHY/LTE_REFSIG/lte_refsig.h"
-
+#include "PHY/sse_intrin.h"
 
 /*!\brief This function generates the NR Gold sequence (38-211, Sec 5.2.1) for the PBCH DMRS.
 @param PHY_VARS_gNB* gNB structure provides configuration, frame parameters and the pointers to the 32 bits sequence storage tables
@@ -49,4 +49,10 @@ int nr_pusch_dmrs_rx(PHY_VARS_gNB *gNB,
                      unsigned char lp,
                      unsigned short nb_pusch_rb,
                      uint8_t dmrs_type);
+
+void init_scrambling_luts(void);
+
+extern __m64 byte2m64_re[256];
+extern __m64 byte2m64_im[256];
+
 #endif
diff --git a/openair1/PHY/NR_TRANSPORT/pucch_rx.c b/openair1/PHY/NR_TRANSPORT/pucch_rx.c
index ee3e247cb0cc1398e43fedbd4af23b71ce9086a0..62ced5af9d28631caa6eea2b886844eb3d117c66 100644
--- a/openair1/PHY/NR_TRANSPORT/pucch_rx.c
+++ b/openair1/PHY/NR_TRANSPORT/pucch_rx.c
@@ -1,3 +1,34 @@
+/*
+ * Licensed to the OpenAirInterface (OAI) Software Alliance under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The OpenAirInterface Software Alliance licenses this file to You under
+ * the OAI Public License, Version 1.1  (the "License"); you may not use this file
+ * except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.openairinterface.org/?page_id=698
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ *-------------------------------------------------------------------------------
+ * For more information about the OpenAirInterface (OAI) Software Alliance:
+ *      contact@openairinterface.org
+ */
+
+/*! \file PHY/NR_TRANSPORT/pucch_rx.c
+* \brief Top-level routines for decoding the PUCCH physical channel
+* \author A. Mico Pereperez, Padarthi Naga Prasanth, Francesco Mani, Raymond Knopp
+* \date 2020
+* \version 0.2
+* \company Eurecom
+* \email:
+* \note
+* \warning
+*/
 #include<stdio.h>
 #include <string.h>
 #include <math.h>
@@ -8,10 +39,12 @@
 
 #include "PHY/impl_defs_nr.h"
 #include "PHY/defs_nr_common.h"
-#include "PHY/defs_nr_UE.h"
+#include "PHY/defs_gNB.h"
+#include "PHY/sse_intrin.h"
 #include "PHY/NR_UE_TRANSPORT/pucch_nr.h"
 #include "PHY/NR_TRANSPORT/nr_transport_common_proto.h"
 #include "PHY/NR_TRANSPORT/nr_transport.h"
+#include "PHY/NR_REFSIG/nr_refsig.h"
 #include "common/utils/LOG/log.h"
 #include "common/utils/LOG/vcd_signal_dumper.h"
 
@@ -188,7 +221,7 @@ void nr_decode_pucch0(PHY_VARS_gNB *gNB,
   for (l=0; l<pucch_pdu->nr_of_symbols; l++) {
 
     l2 = l+pucch_pdu->start_symbol_index;
-    re_offset = (12*pucch_pdu->prb_start) + frame_parms->first_carrier_offset;
+    re_offset = (12*pucch_pdu->prb_start) + (12*pucch_pdu->bwp_start) + frame_parms->first_carrier_offset;
     if (re_offset>= frame_parms->ofdm_symbol_size) 
       re_offset-=frame_parms->ofdm_symbol_size;
 
@@ -818,3 +851,496 @@ void nr_decode_pucch1(  int32_t **rxdataF,
   }
 }
 
+__m256i pucch2_3bit[8*2];
+__m256i pucch2_4bit[16*2];
+__m256i pucch2_5bit[32*2];
+__m256i pucch2_6bit[64*2];
+__m256i pucch2_7bit[128*2];
+__m256i pucch2_8bit[256*2];
+__m256i pucch2_9bit[512*2];
+__m256i pucch2_10bit[1024*2];
+__m256i pucch2_11bit[2048*2];
+
+__m256i *pucch2_lut[9]={pucch2_3bit,
+			pucch2_4bit,
+			pucch2_5bit,
+			pucch2_6bit,
+			pucch2_7bit,
+			pucch2_8bit,
+			pucch2_9bit,
+			pucch2_10bit,
+			pucch2_11bit};
+
+void init_pucch2_luts() {
+
+  uint32_t out;
+  int8_t bit; 
+  
+  for (int b=3;b<12;b++) {
+    for (uint16_t i=0;i<(1<<b);i++) {
+      out=encodeSmallBlock(&i,b);
+      if (b==3) printf("in %d, out %x\n",i,out);
+      __m256i *lut_i=&pucch2_lut[b-3][i<<1];
+      __m256i *lut_ip1=&pucch2_lut[b-3][1+(i<<1)];
+      bit = (out&0x1) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,0);
+      bit = (out&0x2) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,0);
+      bit = (out&0x4) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,1);
+      bit = (out&0x8) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,1);
+      bit = (out&0x10) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,2);
+      bit = (out&0x20) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,2);
+      bit = (out&0x40) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,3);
+      bit = (out&0x80) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,3);
+      bit = (out&0x100) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,4);
+      bit = (out&0x200) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,4);
+      bit = (out&0x400) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,5);
+      bit = (out&0x800) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,5);
+      bit = (out&0x1000) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,6);
+      bit = (out&0x2000) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,6);
+      bit = (out&0x4000) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,7);
+      bit = (out&0x8000) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,7);
+      bit = (out&0x10000) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,8);
+      bit = (out&0x20000) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,8);
+      bit = (out&0x40000) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,9);
+      bit = (out&0x80000) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,9);
+      bit = (out&0x100000) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,10);
+      bit = (out&0x200000) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,10);
+      bit = (out&0x400000) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,11);
+      bit = (out&0x800000) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,11);
+      bit = (out&0x1000000) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,12);
+      bit = (out&0x2000000) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,12);
+      bit = (out&0x4000000) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,13);
+      bit = (out&0x8000000) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,13);
+      bit = (out&0x10000000) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,14);
+      bit = (out&0x20000000) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,14);
+      bit = (out&0x40000000) > 0 ? -1 : 1;
+      *lut_i = _mm256_insert_epi16(*lut_i,bit,15);
+      bit = (out&0x80000000) > 0 ? -1 : 1;
+      *lut_ip1 = _mm256_insert_epi16(*lut_ip1,bit,15);
+    }
+  }
+}
+
+
+void nr_decode_pucch2(PHY_VARS_gNB *gNB,
+                      int slot,
+                      nfapi_nr_uci_pucch_pdu_format_2_3_4_t* uci_pdu,
+                      nfapi_nr_pucch_pdu_t* pucch_pdu) {
+
+  int32_t **rxdataF = gNB->common_vars.rxdataF;
+  NR_DL_FRAME_PARMS *frame_parms = &gNB->frame_parms;
+  pucch_GroupHopping_t pucch_GroupHopping = pucch_pdu->group_hop_flag + (pucch_pdu->sequence_hop_flag<<1);
+
+  AssertFatal(pucch_pdu->nr_of_symbols==1 || pucch_pdu->nr_of_symbols==2,
+	      "Illegal number of symbols  for PUCCH 2 %d\n",pucch_pdu->nr_of_symbols);
+
+
+  //extract pucch and dmrs first
+
+  int l2;
+  int re_offset = (12*pucch_pdu->prb_start) + (12*pucch_pdu->bwp_start) + frame_parms->first_carrier_offset;
+  if (re_offset>= frame_parms->ofdm_symbol_size) 
+    re_offset-=frame_parms->ofdm_symbol_size;
+  
+  AssertFatal(pucch_pdu->prb_size*pucch_pdu->nr_of_symbols > 1,"number of PRB*SYMB (%d,%d)< 2",
+	      pucch_pdu->prb_size,pucch_pdu->nr_of_symbols);
+
+  int Prx = gNB->gNB_config.carrier_config.num_rx_ant.value;
+  int Prx2 = (Prx==1)?2:Prx;
+  // use 2 for Nb antennas in case of single antenna to allow the following allocations
+  int16_t r_re_ext[Prx2][8*pucch_pdu->nr_of_symbols*pucch_pdu->prb_size] __attribute__((aligned(32)));
+  int16_t r_im_ext[Prx2][8*pucch_pdu->nr_of_symbols*pucch_pdu->prb_size] __attribute__((aligned(32)));
+  int16_t r_re_ext2[Prx2][8*pucch_pdu->nr_of_symbols*pucch_pdu->prb_size] __attribute__((aligned(32)));
+  int16_t r_im_ext2[Prx2][8*pucch_pdu->nr_of_symbols*pucch_pdu->prb_size] __attribute__((aligned(32)));
+  int16_t rd_re_ext[Prx2][4*pucch_pdu->nr_of_symbols*pucch_pdu->prb_size] __attribute__((aligned(32)));
+  int16_t rd_im_ext[Prx2][4*pucch_pdu->nr_of_symbols*pucch_pdu->prb_size] __attribute__((aligned(32)));
+
+  int16_t *rp[Prx2];
+  __m64 dmrs_re,dmrs_im;
+
+  for (int aa=0;aa<Prx;aa++) rp[aa] = ((int16_t *)&rxdataF[aa][(l2*frame_parms->ofdm_symbol_size)+re_offset]);
+
+#ifdef DEBUG_NR_PUCCH_RX
+  printf("Decoding pucch2 for %d symbols, %d PRB\n",pucch_pdu->nr_of_symbols,pucch_pdu->prb_size);
+#endif
+
+  int nc_group_size=1; // 2 PRB
+  int ngroup = pucch_pdu->prb_size/nc_group_size/2;
+  int32_t corr32_re[ngroup][Prx2],corr32_im[ngroup][Prx2];
+  for (int aa=0;aa<Prx;aa++) for (int group=0;group<ngroup;group++) { corr32_re[group][aa]=0; corr32_im[group][aa]=0;}
+
+  if (pucch_pdu->nr_of_symbols == 1) {
+     AssertFatal((pucch_pdu->prb_size&1) == 0,"prb_size %d is not a multiple of 2\n",pucch_pdu->prb_size);
+     // 24 PRBs contains 48x16-bit, so 6x8x16-bit 
+     for (int prb=0;prb<pucch_pdu->prb_size;prb+=2) {
+        for (int aa=0;aa<Prx;aa++) {
+
+	  r_re_ext[aa][0]=rp[aa][0];
+	  r_im_ext[aa][0]=rp[aa][1];
+	  rd_re_ext[aa][0]=rp[aa][2];
+	  rd_im_ext[aa][0]=rp[aa][3];
+	  r_re_ext[aa][1]=rp[aa][4];
+	  r_im_ext[aa][1]=rp[aa][5];
+
+	  r_re_ext[aa][2]=rp[aa][6];
+	  r_im_ext[aa][2]=rp[aa][7];
+	  rd_re_ext[aa][1]=rp[aa][8];
+	  rd_im_ext[aa][1]=rp[aa][9];
+	  r_re_ext[aa][3]=rp[aa][10];
+	  r_im_ext[aa][3]=rp[aa][11];
+
+	  r_re_ext[aa][4]=rp[aa][12];
+	  r_im_ext[aa][4]=rp[aa][13];
+	  rd_re_ext[aa][2]=rp[aa][14];
+	  rd_im_ext[aa][2]=rp[aa][15];
+	  r_re_ext[aa][5]=rp[aa][16];
+	  r_im_ext[aa][5]=rp[aa][17];
+
+	  r_re_ext[aa][6]=rp[aa][18];
+	  r_im_ext[aa][6]=rp[aa][19];
+	  rd_re_ext[aa][3]=rp[aa][20];
+	  rd_im_ext[aa][3]=rp[aa][21];
+	  r_re_ext[aa][7]=rp[aa][22];
+	  r_im_ext[aa][7]=rp[aa][23];
+
+	  r_re_ext[aa][8]=rp[aa][24];
+	  r_im_ext[aa][8]=rp[aa][25];
+	  rd_re_ext[aa][4]=rp[aa][26];
+	  rd_im_ext[aa][4]=rp[aa][27];
+	  r_re_ext[aa][9]=rp[aa][28];
+	  r_im_ext[aa][9]=rp[aa][29];
+
+	  r_re_ext[aa][10]=rp[aa][30];
+	  r_im_ext[aa][10]=rp[aa][31];
+	  rd_re_ext[aa][5]=rp[aa][32];
+	  rd_im_ext[aa][5]=rp[aa][33];
+	  r_re_ext[aa][11]=rp[aa][34];
+	  r_im_ext[aa][11]=rp[aa][35];
+
+	  r_re_ext[aa][12]=rp[aa][36];
+	  r_im_ext[aa][12]=rp[aa][37];
+	  rd_re_ext[aa][6]=rp[aa][38];
+	  rd_im_ext[aa][6]=rp[aa][39];
+	  r_re_ext[aa][13]=rp[aa][40];
+	  r_im_ext[aa][13]=rp[aa][41];
+
+	  r_re_ext[aa][14]=rp[aa][42];
+	  r_im_ext[aa][14]=rp[aa][43];
+	  rd_re_ext[aa][7]=rp[aa][44];
+	  rd_im_ext[aa][7]=rp[aa][45];
+	  r_re_ext[aa][15]=rp[aa][46];
+	  r_im_ext[aa][15]=rp[aa][47];
+		  
+#ifdef DEBUG_NR_PUCCH_RX
+	  for (int i=0;i<8;i++) printf("Ant %d PRB %d dmrs[%d] -> (%d,%d)\n",aa,prb+(i>>2),i,rd_re_ext[aa][i],rd_im_ext[aa],i);
+#endif
+	} // aa
+     } // prb
+
+
+     // first compute DMRS component
+     uint32_t x1, x2, s=0;
+     x2 = (((1<<17)*((14*slot) + (pucch_pdu->start_symbol_index) + 1)*((2*pucch_pdu->dmrs_scrambling_id) + 1)) + (2*pucch_pdu->dmrs_scrambling_id))%(1U<<31); // c_init calculation according to TS38.211 subclause
+#ifdef DEBUG_NR_PUCCH_RX
+     printf("slot %d, start_symbol_index %d, dmrs_scrambling_id %d\n",
+       slot,pucch_pdu->start_symbol_index,pucch_pdu->dmrs_scrambling_id);
+#endif
+     s = lte_gold_generic(&x1, &x2, 1);
+
+
+     for (int group=0;group<ngroup;group++) {
+     // each group has 8*nc_group_size elements, compute 1 complex correlation with DMRS per group
+     // non-coherent combining across groups
+       dmrs_re = byte2m64_re[((uint8_t*)&s)[(group&1)<<1]];
+       dmrs_im = byte2m64_im[((uint8_t*)&s)[(group&1)<<1]];
+#ifdef DEBUG_NR_PUCCH_RX
+       printf("Group %d: s %x x2 %x ((%d,%d),(%d,%d),(%d,%d),(%d,%d))\n",
+	      group,
+	      ((uint16_t*)&s)[0],x2,
+	      ((int16_t*)&dmrs_re)[0],((int16_t*)&dmrs_im)[0],    
+	      ((int16_t*)&dmrs_re)[1],((int16_t*)&dmrs_im)[1],    
+	      ((int16_t*)&dmrs_re)[2],((int16_t*)&dmrs_im)[2],    
+	      ((int16_t*)&dmrs_re)[3],((int16_t*)&dmrs_im)[3]);   
+#endif
+       for (int aa=0;aa<Prx;aa++) {
+#ifdef DEBUG_NR_PUCCH_RX
+	 printf("Group %d: rd ((%d,%d),(%d,%d),(%d,%d),(%d,%d))\n",
+		group,
+		rd_re_ext[aa][0],rd_im_ext[aa][0],
+		rd_re_ext[aa][1],rd_im_ext[aa][1],
+		rd_re_ext[aa][2],rd_im_ext[aa][2],
+		rd_re_ext[aa][3],rd_im_ext[aa][3]);
+#endif
+	 corr32_re[group][aa]+=(rd_re_ext[aa][0]*((int16_t*)&dmrs_re)[0] + rd_im_ext[aa][0]*((int16_t*)&dmrs_im)[0]); 
+	 corr32_im[group][aa]+=(-rd_re_ext[aa][0]*((int16_t*)&dmrs_im)[0] + rd_im_ext[aa][0]*((int16_t*)&dmrs_re)[0]); 
+	 corr32_re[group][aa]+=(rd_re_ext[aa][1]*((int16_t*)&dmrs_re)[1] + rd_im_ext[aa][1]*((int16_t*)&dmrs_im)[1]); 
+	 corr32_im[group][aa]+=(-rd_re_ext[aa][1]*((int16_t*)&dmrs_im)[1] + rd_im_ext[aa][1]*((int16_t*)&dmrs_re)[1]); 
+	 corr32_re[group][aa]+=(rd_re_ext[aa][2]*((int16_t*)&dmrs_re)[2] + rd_im_ext[aa][2]*((int16_t*)&dmrs_im)[2]); 
+	 corr32_im[group][aa]+=(-rd_re_ext[aa][2]*((int16_t*)&dmrs_im)[2] + rd_im_ext[aa][2]*((int16_t*)&dmrs_re)[2]); 
+	 corr32_re[group][aa]+=(rd_re_ext[aa][3]*((int16_t*)&dmrs_re)[3] + rd_im_ext[aa][3]*((int16_t*)&dmrs_im)[3]); 
+	 corr32_im[group][aa]+=(-rd_re_ext[aa][3]*((int16_t*)&dmrs_im)[3] + rd_im_ext[aa][3]*((int16_t*)&dmrs_re)[3]); 
+       }
+       dmrs_re = byte2m64_re[((uint8_t*)&s)[1+((group&1)<<1)]];
+       dmrs_im = byte2m64_im[((uint8_t*)&s)[1+((group&1)<<1)]];
+#ifdef DEBUG_NR_PUCCH_RX
+       printf("Group %d: s %x ((%d,%d),(%d,%d),(%d,%d),(%d,%d))\n",
+	      group,
+	      ((uint16_t*)&s)[1],
+	      ((int16_t*)&dmrs_re)[0],((int16_t*)&dmrs_im)[0],    
+	      ((int16_t*)&dmrs_re)[1],((int16_t*)&dmrs_im)[1],    
+	      ((int16_t*)&dmrs_re)[2],((int16_t*)&dmrs_im)[2],    
+	      ((int16_t*)&dmrs_re)[3],((int16_t*)&dmrs_im)[3]);
+#endif
+       for (int aa=0;aa<Prx;aa++) {
+#ifdef DEBUG_NR_PUCCH_RX
+	 printf("Group %d: rd ((%d,%d),(%d,%d),(%d,%d),(%d,%d))\n",
+		group,
+		rd_re_ext[aa][4],rd_im_ext[aa][4],
+		rd_re_ext[aa][5],rd_im_ext[aa][5],
+		rd_re_ext[aa][6],rd_im_ext[aa][6],
+		rd_re_ext[aa][7],rd_im_ext[aa][7]);
+#endif
+	 corr32_re[group][aa]+=(rd_re_ext[aa][4]*((int16_t*)&dmrs_re)[0] + rd_im_ext[aa][4]*((int16_t*)&dmrs_im)[0]); 
+	 corr32_im[group][aa]+=(-rd_re_ext[aa][4]*((int16_t*)&dmrs_im)[0] + rd_im_ext[aa][4]*((int16_t*)&dmrs_re)[0]); 
+	 corr32_re[group][aa]+=(rd_re_ext[aa][5]*((int16_t*)&dmrs_re)[1] + rd_im_ext[aa][5]*((int16_t*)&dmrs_im)[1]); 
+	 corr32_im[group][aa]+=(-rd_re_ext[aa][5]*((int16_t*)&dmrs_im)[1] + rd_im_ext[aa][5]*((int16_t*)&dmrs_re)[1]); 
+	 corr32_re[group][aa]+=(rd_re_ext[aa][6]*((int16_t*)&dmrs_re)[2] + rd_im_ext[aa][6]*((int16_t*)&dmrs_im)[2]); 
+	 corr32_im[group][aa]+=(-rd_re_ext[aa][6]*((int16_t*)&dmrs_im)[2] + rd_im_ext[aa][6]*((int16_t*)&dmrs_re)[2]); 
+	 corr32_re[group][aa]+=(rd_re_ext[aa][7]*((int16_t*)&dmrs_re)[3] + rd_im_ext[aa][7]*((int16_t*)&dmrs_im)[3]); 
+	 corr32_im[group][aa]+=(-rd_re_ext[aa][7]*((int16_t*)&dmrs_im)[3] + rd_im_ext[aa][7]*((int16_t*)&dmrs_re)[3]); 
+	 corr32_re[group][aa]>>=5;
+	 corr32_im[group][aa]>>=5;
+#ifdef DEBUG_NR_PUCCH_RX
+	 printf("Group %d: corr32 (%d,%d)\n",group,corr32_re[group][aa],corr32_im[group][aa]);
+#endif
+       } //aa    
+       
+       if ((group&3) == 3) s = lte_gold_generic(&x1, &x2, 0);
+     } // group
+  }
+  else { // 2 symbol case
+     AssertFatal(1==0, "Fill in 2 symbol PUCCH2 case\n");
+  }
+
+  uint32_t x1, x2, s=0;  
+  // unscrambling
+  x2 = ((pucch_pdu->rnti)<<15)+pucch_pdu->data_scrambling_id;
+  s = lte_gold_generic(&x1, &x2, 1);
+#ifdef DEBUG_NR_PUCCH_RX
+  printf("x2 %x, s %x\n",x2,s);
+#endif
+  __m64 c_re0,c_im0,c_re1,c_im1,c_re2,c_im2,c_re3,c_im3;
+  re_offset=0;
+  for (int prb=0;prb<pucch_pdu->prb_size;prb+=2,re_offset+=16) {
+    c_re0 = byte2m64_re[((uint8_t*)&s)[0]];
+    c_im0 = byte2m64_im[((uint8_t*)&s)[0]];
+    c_re1 = byte2m64_re[((uint8_t*)&s)[1]];
+    c_im1 = byte2m64_im[((uint8_t*)&s)[1]];
+    c_re2 = byte2m64_re[((uint8_t*)&s)[2]];
+    c_im2 = byte2m64_im[((uint8_t*)&s)[2]];
+    c_re3 = byte2m64_re[((uint8_t*)&s)[3]];
+    c_im3 = byte2m64_im[((uint8_t*)&s)[3]];
+
+    for (int aa=0;aa<Prx;aa++) {
+#ifdef DEBUG_NR_PUCCH_RX
+      printf("prb %d: rd ((%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d))\n",
+		prb,
+		r_re_ext[aa][re_offset],r_im_ext[aa][re_offset],
+		r_re_ext[aa][re_offset+1],r_im_ext[aa][re_offset+1],
+		r_re_ext[aa][re_offset+2],r_im_ext[aa][re_offset+2],
+		r_re_ext[aa][re_offset+3],r_im_ext[aa][re_offset+3],
+		r_re_ext[aa][re_offset+4],r_im_ext[aa][re_offset+4],
+		r_re_ext[aa][re_offset+5],r_im_ext[aa][re_offset+5],
+		r_re_ext[aa][re_offset+6],r_im_ext[aa][re_offset+6],
+		r_re_ext[aa][re_offset+7],r_im_ext[aa][re_offset+7]);
+      printf("prb %d: c ((%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d))\n",
+		prb,
+		((int16_t*)&c_re0)[0],((int16_t*)&c_im0)[0],
+		((int16_t*)&c_re0)[1],((int16_t*)&c_im0)[1],
+		((int16_t*)&c_re0)[2],((int16_t*)&c_im0)[2],
+		((int16_t*)&c_re0)[3],((int16_t*)&c_im0)[3],
+		((int16_t*)&c_re1)[0],((int16_t*)&c_im1)[0],
+		((int16_t*)&c_re1)[1],((int16_t*)&c_im1)[1],
+		((int16_t*)&c_re1)[2],((int16_t*)&c_im1)[2],
+		((int16_t*)&c_re1)[3],((int16_t*)&c_im1)[3]
+		);
+      printf("prb %d: rd ((%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d))\n",
+		prb+1,
+		r_re_ext[aa][re_offset+8],r_im_ext[aa][re_offset+8],
+		r_re_ext[aa][re_offset+9],r_im_ext[aa][re_offset+9],
+		r_re_ext[aa][re_offset+10],r_im_ext[aa][re_offset+10],
+		r_re_ext[aa][re_offset+11],r_im_ext[aa][re_offset+11],
+		r_re_ext[aa][re_offset+12],r_im_ext[aa][re_offset+12],
+		r_re_ext[aa][re_offset+13],r_im_ext[aa][re_offset+13],
+		r_re_ext[aa][re_offset+14],r_im_ext[aa][re_offset+14],
+		r_re_ext[aa][re_offset+15],r_im_ext[aa][re_offset+15]);
+      printf("prb %d: c ((%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d))\n",
+		prb+1,
+		((int16_t*)&c_re2)[0],((int16_t*)&c_im2)[0],
+		((int16_t*)&c_re2)[1],((int16_t*)&c_im2)[1],
+		((int16_t*)&c_re2)[2],((int16_t*)&c_im2)[2],
+		((int16_t*)&c_re2)[3],((int16_t*)&c_im2)[3],
+		((int16_t*)&c_re3)[0],((int16_t*)&c_im3)[0],
+		((int16_t*)&c_re3)[1],((int16_t*)&c_im3)[1],
+		((int16_t*)&c_re3)[2],((int16_t*)&c_im3)[2],
+		((int16_t*)&c_re3)[3],((int16_t*)&c_im3)[3]
+		);
+#endif
+
+      ((__m64*)&r_re_ext2[aa][re_offset])[0] = _mm_mullo_pi16(((__m64*)&r_re_ext[aa][re_offset])[0],c_im0);
+      ((__m64*)&r_re_ext[aa][re_offset])[0] = _mm_mullo_pi16(((__m64*)&r_re_ext[aa][re_offset])[0],c_re0);
+      ((__m64*)&r_im_ext2[aa][re_offset])[0] = _mm_mullo_pi16(((__m64*)&r_im_ext[aa][re_offset])[0],c_re0);
+      ((__m64*)&r_im_ext[aa][re_offset])[0] = _mm_mullo_pi16(((__m64*)&r_im_ext[aa][re_offset])[0],c_im0);
+
+      ((__m64*)&r_re_ext2[aa][re_offset])[1] = _mm_mullo_pi16(((__m64*)&r_re_ext[aa][re_offset])[1],c_im1);
+      ((__m64*)&r_re_ext[aa][re_offset])[1] = _mm_mullo_pi16(((__m64*)&r_re_ext[aa][re_offset])[1],c_re1);
+      ((__m64*)&r_im_ext2[aa][re_offset])[1] = _mm_mullo_pi16(((__m64*)&r_im_ext[aa][re_offset])[1],c_re1);
+      ((__m64*)&r_im_ext[aa][re_offset])[1] = _mm_mullo_pi16(((__m64*)&r_im_ext[aa][re_offset])[1],c_im1);
+
+      ((__m64*)&r_re_ext2[aa][re_offset])[2] = _mm_mullo_pi16(((__m64*)&r_re_ext[aa][re_offset])[2],c_im2);
+      ((__m64*)&r_re_ext[aa][re_offset])[2] = _mm_mullo_pi16(((__m64*)&r_re_ext[aa][re_offset])[2],c_re2);
+      ((__m64*)&r_im_ext2[aa][re_offset])[2] = _mm_mullo_pi16(((__m64*)&r_im_ext[aa][re_offset])[2],c_re2);
+      ((__m64*)&r_im_ext[aa][re_offset])[2] = _mm_mullo_pi16(((__m64*)&r_im_ext[aa][re_offset])[2],c_im2);
+
+      ((__m64*)&r_re_ext2[aa][re_offset])[3] = _mm_mullo_pi16(((__m64*)&r_re_ext[aa][re_offset])[3],c_im3);
+      ((__m64*)&r_re_ext[aa][re_offset])[3] = _mm_mullo_pi16(((__m64*)&r_re_ext[aa][re_offset])[3],c_re3);
+      ((__m64*)&r_im_ext2[aa][re_offset])[3] = _mm_mullo_pi16(((__m64*)&r_im_ext[aa][re_offset])[3],c_re3);
+      ((__m64*)&r_im_ext[aa][re_offset])[3] = _mm_mullo_pi16(((__m64*)&r_im_ext[aa][re_offset])[3],c_im3);
+
+#ifdef DEBUG_NR_PUCCH_RX
+      printf("prb %d: r ((%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d))\n",
+	     prb,
+	     r_re_ext[aa][re_offset],r_im_ext[aa][re_offset],
+	     r_re_ext[aa][re_offset+1],r_im_ext[aa][re_offset+1],
+	     r_re_ext[aa][re_offset+2],r_im_ext[aa][re_offset+2],
+	     r_re_ext[aa][re_offset+3],r_im_ext[aa][re_offset+3],
+	     r_re_ext[aa][re_offset+4],r_im_ext[aa][re_offset+4],
+	     r_re_ext[aa][re_offset+5],r_im_ext[aa][re_offset+5],
+	     r_re_ext[aa][re_offset+6],r_im_ext[aa][re_offset+6],
+	     r_re_ext[aa][re_offset+7],r_im_ext[aa][re_offset+7]);
+      printf("prb %d: r ((%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d),(%d,%d))\n",
+	     prb+1,
+	     r_re_ext[aa][re_offset+8],r_im_ext[aa][re_offset+8],
+	     r_re_ext[aa][re_offset+9],r_im_ext[aa][re_offset+9],
+	     r_re_ext[aa][re_offset+10],r_im_ext[aa][re_offset+10],
+	     r_re_ext[aa][re_offset+11],r_im_ext[aa][re_offset+11],
+	     r_re_ext[aa][re_offset+12],r_im_ext[aa][re_offset+12],
+	     r_re_ext[aa][re_offset+13],r_im_ext[aa][re_offset+13],
+	     r_re_ext[aa][re_offset+14],r_im_ext[aa][re_offset+14],
+	     r_re_ext[aa][re_offset+15],r_im_ext[aa][re_offset+15]);
+#endif      
+    }
+    s = lte_gold_generic(&x1, &x2, 0);
+  }
+  AssertFatal(pucch_pdu->bit_len_csi_part1 + pucch_pdu->bit_len_csi_part2 == 0,"no csi for now\n");
+  AssertFatal((pucch_pdu->bit_len_harq+pucch_pdu->sr_flag > 2 ) && (pucch_pdu->bit_len_harq+pucch_pdu->sr_flag < 12),"illegal length (%d,%d)\n",pucch_pdu->bit_len_harq,pucch_pdu->sr_flag);
+  int nb_bit = pucch_pdu->bit_len_harq+pucch_pdu->sr_flag;
+  __m256i *rp_re[Prx2];
+  __m256i *rp2_re[Prx2];
+  __m256i *rp_im[Prx2];
+  __m256i *rp2_im[Prx2];
+  for (int aa=0;aa<Prx;aa++) {
+    rp_re[aa] = (__m256i*)r_re_ext[aa];
+    rp_im[aa] = (__m256i*)r_im_ext[aa];
+    rp2_re[aa] = (__m256i*)r_re_ext2[aa];
+    rp2_im[aa] = (__m256i*)r_im_ext2[aa];
+  }
+  __m256i prod_re[Prx2],prod_im[Prx2];
+  int64_t corr=0;
+  int cw_ML=0;
+
+  for (int cw=0;cw<1<<nb_bit;cw++) {
+#ifdef DEBUG_NR_PUCCH_RX
+    printf("cw %d:",cw);
+    for (int i=0;i<32;i+=2) {
+      printf("%d,%d,",
+	     ((int16_t*)&pucch2_lut[nb_bit-3][cw<<1])[i>>1],
+	     ((int16_t*)&pucch2_lut[nb_bit-3][cw<<1])[1+(i>>1)]);
+    }
+    printf("\n");
+#endif
+    // do complex correlation
+    for (int aa=0;aa<Prx;aa++) {
+      prod_re[aa] = _mm256_srai_epi16(_mm256_adds_epi16(_mm256_mullo_epi16(pucch2_lut[nb_bit-3][cw<<1],rp_re[aa][0]),
+							_mm256_mullo_epi16(pucch2_lut[nb_bit-3][(cw<<1)+1],rp_im[aa][0])),5);
+      prod_im[aa] = _mm256_srai_epi16(_mm256_subs_epi16(_mm256_mullo_epi16(pucch2_lut[nb_bit-3][cw<<1],rp2_im[aa][0]),
+							_mm256_mullo_epi16(pucch2_lut[nb_bit-3][(cw<<1)+1],rp2_re[aa][0])),5);
+      prod_re[aa] = _mm256_hadds_epi16(prod_re[aa],prod_re[aa]);// 0+1
+      prod_im[aa] = _mm256_hadds_epi16(prod_im[aa],prod_im[aa]);
+      prod_re[aa] = _mm256_hadds_epi16(prod_re[aa],prod_re[aa]);// 0+1+2+3
+      prod_im[aa] = _mm256_hadds_epi16(prod_im[aa],prod_im[aa]);
+      prod_re[aa] = _mm256_hadds_epi16(prod_re[aa],prod_re[aa]);// 0+1+2+3+4+5+6+7
+      prod_im[aa] = _mm256_hadds_epi16(prod_im[aa],prod_im[aa]);
+      prod_re[aa] = _mm256_hadds_epi16(prod_re[aa],prod_re[aa]);// 0+1+2+3+4+5+6+7+8+9+10+11+12+13+14+15
+      prod_im[aa] = _mm256_hadds_epi16(prod_im[aa],prod_im[aa]);
+    }
+    int64_t corr_re=0,corr_im=0;
+
+    for (int aa=0;aa<Prx;aa++) {
+      LOG_D(PHY,"pucch2 cw %d aa %d: (%d,%d)+(%d,%d) = (%d,%d)\n",cw,aa,
+	    corr32_re[0][aa],corr32_im[0][aa],
+	    ((int16_t*)(&prod_re[aa]))[0],
+	    ((int16_t*)(&prod_im[aa]))[0],
+	    corr32_re[0][aa]+((int16_t*)(&prod_re[0]))[0],
+	    corr32_im[0][aa]+((int16_t*)(&prod_im[0]))[0]);
+	     
+      corr_re += ( corr32_re[0][aa]+((int16_t*)(&prod_re[0]))[0]);
+      corr_im += ( corr32_im[0][aa]+((int16_t*)(&prod_im[0]))[0]);
+
+    }
+    int64_t corr_tmp = corr_re*corr_re + corr_im*corr_im;
+    if (corr_tmp > corr) {
+      corr = corr_tmp;
+      cw_ML=cw;
+    }
+  }
+  uint8_t corr_dB = dB_fixed64((uint64_t)corr);
+  LOG_D(PHY,"cw_ML %d, metric %d dB\n",cw_ML,corr_dB);
+  uci_pdu->harq.harq_bit_len = pucch_pdu->bit_len_harq;
+
+  
+  int harq_bytes=pucch_pdu->bit_len_harq>>3;
+  if ((pucch_pdu->bit_len_harq&7) > 0) harq_bytes++;
+  uci_pdu->harq.harq_payload = (nfapi_nr_harq_t*)malloc(harq_bytes);
+  uci_pdu->harq.harq_crc = 2;
+  for (int i=0;i<harq_bytes;i++) {
+    uci_pdu->harq.harq_payload[i] = cw_ML & 255;
+    cw_ML>>=8;
+  }
+  
+  if (pucch_pdu->sr_flag == 1) {
+    uci_pdu->sr.sr_bit_len = 1;
+    uci_pdu->sr.sr_payload = malloc(1);
+    uci_pdu->sr.sr_payload[0] = cw_ML;
+  }
+}
+    
diff --git a/openair1/PHY/NR_UE_TRANSPORT/pucch_nr.c b/openair1/PHY/NR_UE_TRANSPORT/pucch_nr.c
index d63cef508539fcb1474ea5b7baee01abb3ad3bf8..2a08d54c5ab141dbddc62719abbbe08b404c12ce 100644
--- a/openair1/PHY/NR_UE_TRANSPORT/pucch_nr.c
+++ b/openair1/PHY/NR_UE_TRANSPORT/pucch_nr.c
@@ -42,6 +42,7 @@
 #include "common/utils/LOG/vcd_signal_dumper.h"
 
 #include "T.h"
+//#define NR_UNIT_TEST 1
 #ifdef NR_UNIT_TEST
   #define DEBUG_PUCCH_TX
   #define DEBUG_NR_PUCCH_TX
@@ -905,7 +906,9 @@ void nr_uci_encoding(uint64_t payload,
   if (A<=11) {
     // procedure in subclause 6.3.1.2.2 (UCI encoded by channel coding of small block lengths -> subclause 6.3.1.3.2)
     // CRC bits are not attached, and coding small block lengths (subclause 5.3.3)
+    b[0] = encodeSmallBlock((uint16_t*)&payload,A);
   } else if (A>=12) {
+    AssertFatal(1==0,"Polar encoding not supported yet for UCI\n");
     // procedure in subclause 6.3.1.2.1 (UCI encoded by Polar code -> subclause 6.3.1.3.1)
     /*if ((A>=360 && E>=1088)||(A>=1013)) {
       I_seg = 1;
@@ -924,10 +927,13 @@ void nr_uci_encoding(uint64_t payload,
     // code block segmentation and CRC attachment is performed according to subclause 5.2.1
     // polar coding subclause 5.3.1
   }
+  
 }
 //#if 0
 void nr_generate_pucch2(PHY_VARS_NR_UE *ue,
                         uint16_t crnti,
+			uint32_t dmrs_scrambling_id,
+			uint32_t data_scrambling_id,
                         int32_t **txdataF,
                         NR_DL_FRAME_PARMS *frame_parms,
                         PUCCH_CONFIG_DEDICATED *pucch_config_dedicated,
@@ -961,14 +967,14 @@ void nr_generate_pucch2(PHY_VARS_NR_UE *ue,
    */
   uint8_t *btilde = malloc(sizeof(int8_t)*M_bit);
   // rnti is given by the C-RNTI
-  uint16_t rnti=crnti, n_id=0;
+  uint16_t rnti=crnti;
 #ifdef DEBUG_NR_PUCCH_TX
   printf("\t [nr_generate_pucch2] rnti = %d ,\n",rnti);
 #endif
   /*
    * Implementing TS 38.211 Subclause 6.3.2.5.1 scrambling format 2
    */
-  nr_pucch2_3_4_scrambling(M_bit,rnti,n_id,b,btilde);
+  nr_pucch2_3_4_scrambling(M_bit,rnti,data_scrambling_id,b,btilde);
   /*
    * Implementing TS 38.211 Subclause 6.3.2.5.2 modulation format 2
    * btilde shall be modulated as described in subclause 5.1 using QPSK
@@ -1016,10 +1022,10 @@ void nr_generate_pucch2(PHY_VARS_NR_UE *ue,
   int m=0;
 
   for (int l=0; l<nrofSymbols; l++) {
-    x2 = (((1<<17)*((14*nr_tti_tx) + (l+startingSymbolIndex) + 1)*((2*n_id) + 1)) + (2*n_id))%(1U<<31); // c_init calculation according to TS38.211 subclause
+    x2 = (((1<<17)*((14*nr_tti_tx) + (l+startingSymbolIndex) + 1)*((2*dmrs_scrambling_id) + 1)) + (2*dmrs_scrambling_id))%(1U<<31); // c_init calculation according to TS38.211 subclause
+
     s = lte_gold_generic(&x1, &x2, 1);
     m = 0;
-
     for (int rb=0; rb<nrofPRB; rb++) {
       //startingPRB = startingPRB + rb;
       if (((rb+startingPRB) <  (frame_parms->N_RB_DL>>1)) && ((frame_parms->N_RB_DL & 1) == 0)) { // if number RBs in bandwidth is even and current PRB is lower band
diff --git a/openair1/PHY/NR_UE_TRANSPORT/pucch_nr.h b/openair1/PHY/NR_UE_TRANSPORT/pucch_nr.h
index 4c4847d7dec911451fe36eba8a5f4fdaf658060b..510abb8373838d75bbcb0eb389c4eb2ef6f629f9 100644
--- a/openair1/PHY/NR_UE_TRANSPORT/pucch_nr.h
+++ b/openair1/PHY/NR_UE_TRANSPORT/pucch_nr.h
@@ -20,7 +20,7 @@
  */
 
 /*! \file PHY/NR_UE_TRANSPORT/pucch_nr.c
-* \brief Top-level routines for generating and decoding the PUCCH physical channel
+* \brief Top-level routines for generating the PUCCH physical channel
 * \author A. Mico Pereperez
 * \date 2018
 * \version 0.1
@@ -70,6 +70,8 @@ void nr_generate_pucch1(PHY_VARS_NR_UE *ue,
                         uint8_t nr_bit);
 void nr_generate_pucch2(PHY_VARS_NR_UE *ue,
                         uint16_t crnti,
+			uint32_t dmrs_scrambling_id,
+			uint32_t data_scrambling_id,
                         int32_t **txdataF,
                         NR_DL_FRAME_PARMS *frame_parms,
                         PUCCH_CONFIG_DEDICATED *pucch_config_dedicated,
diff --git a/openair1/SIMULATION/NR_PHY/pucchsim.c b/openair1/SIMULATION/NR_PHY/pucchsim.c
index 3b7305bf9bf6e281c382e542f8458a951f4668d5..9e1609c3b4c674eeab0f52ea5de05c4720cfc6b9 100644
--- a/openair1/SIMULATION/NR_PHY/pucchsim.c
+++ b/openair1/SIMULATION/NR_PHY/pucchsim.c
@@ -93,11 +93,12 @@ int main(int argc, char **argv)
   uint8_t nrofSymbols=1; //number of OFDM symbols can be 1-2 for format 1
   uint8_t startingSymbolIndex=0; // resource allocated see 9.2.1, 38.213 for more info.should be actually present in the resource set provided
   uint16_t startingPRB=0,startingPRB_intraSlotHopping=0; //PRB number not sure see 9.2.1, 38.213 for more info. Should be actually present in the resource set provided
+  uint16_t nrofPRB=2;
   uint8_t timeDomainOCC=0;
   SCM_t channel_model=AWGN;//Rayleigh1_anticorr;
   
   int N_RB_DL=273,mu=1;
-  float target_error_rate=0.01;
+  float target_error_rate=0.001;
   int frame_length_complex_samples;
   //int frame_length_complex_samples_no_prefix;
   NR_DL_FRAME_PARMS *frame_parms;
@@ -112,9 +113,8 @@ int main(int argc, char **argv)
 
   randominit(0);
   logInit();
-  set_glog(loglvl);
 
-  while ((c = getopt (argc, argv, "f:hA:f:g:i:I:P:B:b:T:m:n:r:o:s:S:x:y:z:N:F:GR:IL")) != -1) {
+  while ((c = getopt (argc, argv, "f:hA:f:g:i:I:P:B:b:T:m:n:r:o:s:S:x:y:z:N:F:GR:IL:q:")) != -1) {
     switch (c) {
     case 'f':
       //write_output_file=1;
@@ -263,6 +263,9 @@ int main(int argc, char **argv)
     case 'r':
       startingPRB=atoi(optarg);
       break;
+    case 'q':
+      nrofPRB=atoi(optarg);
+      break;
     case 'P':
       format=atoi(optarg);
       break;
@@ -305,7 +308,8 @@ int main(int argc, char **argv)
       printf("-F Input filename (.txt format) for RX conformance testing\n");
       printf("-i Enter number of ofdm symbols for pucch\n");
       printf("-I Starting symbol index for pucch\n");
-      printf("-r PUCCCH starting PRB\n");
+      printf("-r PUCCH starting PRB\n");
+      printf("-q PUCCH number of PRB\n");
       printf("-P Enter the format of PUCCH\n");
       printf("-b number of HARQ bits (1-2)\n");
       printf("-B payload to be transmitted on PUCCH\n");
@@ -316,18 +320,22 @@ int main(int argc, char **argv)
     }
   }
 
+  set_glog(loglvl);
+
   if (snr1set==0) snr1 = snr0+10;
 
   printf("Initializing gNodeB for mu %d, N_RB_DL %d\n",mu,N_RB_DL);
 
-  if((format!=0) && (format!=1)){
+  if((format!=0) && (format!=1) && (format!=2)){
     printf("PUCCH format %d not supported\n",format);
     exit(0); 
   }
 
-  AssertFatal((nr_bit<3)&&(actual_payload<4),"Only format 0 and 1 currently supported. 2 bits max supported");
+  AssertFatal(((format < 2)&&(nr_bit<3)&&(actual_payload<4)) ||
+	      ((format == 2)&&(nr_bit>2)&&(nr_bit<12)),"illegal combination format %d, nr_bit %d\n",
+	      format,nr_bit);
 
-  if (nr_bit==1) actual_payload = actual_payload&1;
+  actual_payload &= ((1<<nr_bit)-1);
 
   printf("Transmitted payload is %ld\n",actual_payload);
 
@@ -417,9 +425,9 @@ int main(int argc, char **argv)
     bzero(r_im[i],frame_length_complex_samples*sizeof(double));
 
     printf("Allocating %d samples for txdataF/rxdataF\n",14*frame_parms->ofdm_symbol_size);
-    txdataF[i] = malloc(14*frame_parms->ofdm_symbol_size*sizeof(int));
+    txdataF[i] = memalign(32,14*frame_parms->ofdm_symbol_size*sizeof(int));
     bzero(txdataF[i],14*frame_parms->ofdm_symbol_size*sizeof(int));
-    rxdataF[i] = malloc(14*frame_parms->ofdm_symbol_size*sizeof(int));
+    rxdataF[i] = memalign(32,14*frame_parms->ofdm_symbol_size*sizeof(int));
     bzero(rxdataF[i],14*frame_parms->ofdm_symbol_size*sizeof(int));
   }
 
@@ -443,8 +451,8 @@ int main(int argc, char **argv)
   uint8_t mcs=0;
   startingPRB_intraSlotHopping=N_RB_DL-1;
   pucch_GroupHopping_t PUCCH_GroupHopping=UE->pucch_config_common_nr->pucch_GroupHopping;
-  uint32_t n_id=UE->pucch_config_common_nr->hoppingId;
-
+  uint32_t hopping_id=UE->pucch_config_common_nr->hoppingId;
+  uint32_t dmrs_scrambling_id = 0, data_scrambling_id=0;
   if(format==0){
   // for now we are not considering SR just HARQ-ACK
     if(nr_bit==1)
@@ -461,9 +469,12 @@ int main(int argc, char **argv)
       if(format==0){
         nr_generate_pucch0(UE,txdataF,frame_parms,UE->pucch_config_dedicated,amp,nr_tti_tx,m0,mcs,nrofSymbols,startingSymbolIndex,startingPRB);
       }
-      else{
+      else if (format == 1){
         nr_generate_pucch1(UE,txdataF,frame_parms,UE->pucch_config_dedicated,actual_payload,amp,nr_tti_tx,m0,nrofSymbols,startingSymbolIndex,startingPRB,startingPRB_intraSlotHopping,0,nr_bit);	
       }
+      else {
+	nr_generate_pucch2(UE,0x1234,dmrs_scrambling_id,data_scrambling_id,txdataF,frame_parms,UE->pucch_config_dedicated,actual_payload,amp,nr_tti_tx,nrofSymbols,startingSymbolIndex,nrofPRB,startingPRB,nr_bit);	
+      }
       
       int txlev = signal_energy(&txdataF[aa][startingSymbolIndex*frame_parms->ofdm_symbol_size],
 				frame_parms->ofdm_symbol_size);
@@ -489,7 +500,7 @@ int main(int argc, char **argv)
 	pucch_pdu.bit_len_harq          = nr_bit;
 	pucch_pdu.sr_flag               = 0;
 	pucch_pdu.nr_of_symbols         = nrofSymbols;
-	pucch_pdu.hopping_id            = n_id;
+	pucch_pdu.hopping_id            = hopping_id;
 	pucch_pdu.initial_cyclic_shift  = 0;
 	pucch_pdu.start_symbol_index    = startingSymbolIndex;
 	pucch_pdu.prb_start             = startingPRB;
@@ -497,21 +508,49 @@ int main(int argc, char **argv)
         if(nr_bit==1)
           ack_nack_errors+=(actual_payload^uci_pdu.harq->harq_list[0].harq_value);
         else
-          ack_nack_errors+=((actual_payload^uci_pdu.harq->harq_list[0].harq_value)+((actual_payload>>1)^uci_pdu.harq->harq_list[1].harq_value));
+          ack_nack_errors+=(((actual_payload&1)^uci_pdu.harq->harq_list[0].harq_value)+((actual_payload>>1)^uci_pdu.harq->harq_list[1].harq_value));
 	free(uci_pdu.harq->harq_list);
       }
-      else{
+      else if (format==1) {
 	
-        nr_decode_pucch1(rxdataF,PUCCH_GroupHopping,n_id,&(payload_received),frame_parms,amp,nr_tti_tx,m0,nrofSymbols,startingSymbolIndex,startingPRB,startingPRB_intraSlotHopping,timeDomainOCC,nr_bit);
+        nr_decode_pucch1(rxdataF,PUCCH_GroupHopping,hopping_id,&(payload_received),frame_parms,amp,nr_tti_tx,m0,nrofSymbols,startingSymbolIndex,startingPRB,startingPRB_intraSlotHopping,timeDomainOCC,nr_bit);
         if(nr_bit==1)
           ack_nack_errors+=((actual_payload^payload_received)&1);
         else
           ack_nack_errors+=((actual_payload^payload_received)&1) + (((actual_payload^payload_received)&2)>>1);
       }
+      else if (format==2) {
+	nfapi_nr_uci_pucch_pdu_format_2_3_4_t uci_pdu;
+	nfapi_nr_pucch_pdu_t pucch_pdu;
+	pucch_pdu.rnti = 0x1234;
+	pucch_pdu.subcarrier_spacing    = 1;
+	pucch_pdu.group_hop_flag        = PUCCH_GroupHopping&1;
+	pucch_pdu.sequence_hop_flag     = (PUCCH_GroupHopping>>1)&1;
+	pucch_pdu.bit_len_harq          = nr_bit;
+	pucch_pdu.sr_flag               = 0;
+	pucch_pdu.nr_of_symbols         = nrofSymbols;
+	pucch_pdu.hopping_id            = hopping_id;
+	pucch_pdu.initial_cyclic_shift  = 0;
+	pucch_pdu.start_symbol_index    = startingSymbolIndex;
+	pucch_pdu.prb_size              = nrofPRB;
+	pucch_pdu.prb_start             = startingPRB;
+	pucch_pdu.dmrs_scrambling_id    = dmrs_scrambling_id;
+	pucch_pdu.data_scrambling_id    = data_scrambling_id;
+        nr_decode_pucch2(gNB,nr_tti_tx,&uci_pdu,&pucch_pdu);
+	int harq_bytes=pucch_pdu.bit_len_harq>>3;
+	if ((pucch_pdu.bit_len_harq&7) > 0) harq_bytes++;
+	for (int i=0;i<harq_bytes;i++)
+	  if (uci_pdu.harq.harq_payload[i] != ((int8_t*)&actual_payload)[i]) {
+	    ack_nack_errors++;
+	    break;
+	  }
+	free(uci_pdu.harq.harq_payload);
+
+      }
       n_errors=((actual_payload^payload_received)&1)+(((actual_payload^payload_received)&2)>>1)+(((actual_payload^payload_received)&4)>>2)+n_errors;
     }
     printf("SNR=%f, n_trials=%d, n_bit_errors=%d\n",SNR,n_trials,ack_nack_errors);
-    if((float)ack_nack_errors/(float)(nr_bit*n_trials)<=target_error_rate){
+    if((float)ack_nack_errors/(float)(n_trials)<=target_error_rate){
       printf("PUCCH test OK\n");
       break;
     }