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

fixing "making SFN dependent on GPS time"

Browse Source 
in fact it syncs the USRP time to GPS time when available and then derives the SFN from the USRP time.
a side effect is that now SFN does not necessarily start at 0 and this also required other changes
 - initialization of the future_ul_tti_request.
 - fixing a bug in get_samples_per_slot
This commit is contained in:
Florian Kaltenberger
2022-04-06 17:40:14 +02:00
parent 5b22d94d37
commit 367ad780a4
4 changed files with 37 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files
openair2/LAYER2/NR_MAC_gNB/gNB_scheduler.c
+2 -2
View File
@@ -69,7 +69,7 @@ void clear_nr_nfapi_information(gNB_MAC_INST * gNB,
nfapi_nr_dl_tti_request_t *DL_req = &gNB->DL_req[0];
nfapi_nr_dl_tti_pdcch_pdu_rel15_t ***pdcch = (nfapi_nr_dl_tti_pdcch_pdu_rel15_t ***)gNB->pdcch_pdu_idx[CC_idP];
nfapi_nr_ul_tti_request_t *future_ul_tti_req =
&gNB->UL_tti_req_ahead[CC_idP][(slotP + num_slots - 1) % num_slots];
&gNB->UL_tti_req_ahead[CC_idP][(slotP+6)]; // + num_slots - 1) % num_slots];
nfapi_nr_ul_dci_request_t *UL_dci_req = &gNB->UL_dci_req[0];
nfapi_nr_tx_data_request_t *TX_req = &gNB->TX_req[0];
@@ -87,7 +87,7 @@ void clear_nr_nfapi_information(gNB_MAC_INST * gNB,
UL_dci_req[CC_idP].numPdus = 0;
/* advance last round's future UL_tti_req to be ahead of current frame/slot */
future_ul_tti_req->SFN = (slotP == 0 ? frameP : frameP + 1) % 1024;
future_ul_tti_req->SFN = ((slotP+6<num_slots) ? frameP : frameP + 1) % 1024;
LOG_D(MAC,"Future_ul_tti SFN = %d for slot %d \n", future_ul_tti_req->SFN, (slotP + num_slots - 1) % num_slots);
/* future_ul_tti_req->Slot is fixed! */
future_ul_tti_req->n_pdus = 0;