maki/openairinterface5g
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge remote-tracking branch 'origin/bugfix-310-IQrecplay' into develop

Browse Source
This commit is contained in:
Cedric Roux
2018-06-18 16:25:48 +02:00
parent d579a74e54 acbd1f1f7d
commit 0c445bedda
6 changed files with 282 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files
openair1/PHY/LTE_TRANSPORT/pucch.c
+3 -16
View File
@@ -795,25 +795,11 @@ uint32_t rx_pucch(PHY_VARS_eNB *eNB,
for (j=0;j<NUMBER_OF_UE_MAX;j++) {
eNB->pucch1_stats_cnt[j][i]=0;
eNB->pucch1ab_stats_cnt[j][i]=0;
}
}
#if defined(USRP_REC_PLAY)
// It's probably bad to do this statically only once.
// Looks like the above is incomplete.
// Such reset needs to be done once a UE PHY structure is being used/re-used
// Don't know if this is ever possible in current architecture
for (i=0;i<10240;i++) {
for (j=0;j<NUMBER_OF_UE_MAX;j++) {
eNB->pucch1_stats[j][i]=0;
#if defined(USRP_REC_PLAY) // not 100% sure
eNB->pucch1_stats_thres[j][i]=0;
#endif
}
}
for (i=0;i<20480;i++) {
for (j=0;j<NUMBER_OF_UE_MAX;j++) {
eNB->pucch1ab_stats[j][i]=0;
}
}
#endif
first_call=0;
}
@@ -1236,6 +1222,7 @@ uint32_t rx_pucch(PHY_VARS_eNB *eNB,
#if defined(USRP_REC_PLAY)
// It looks like the value is a bit messy when RF is replayed.
// For instance i assume to skip pucch1_thres from the test below.
// Not 100% sure
if (sigma2_dB<(dB_fixed(stat_max))) {//
#else
if (sigma2_dB<(dB_fixed(stat_max)-pucch1_thres)) {//