Adding second layer timing advance
- Renamed position coordinates function (better readability) - Introduced large timing advance compensation mechanism
This commit is contained in:
@@ -394,7 +394,6 @@ typedef struct PHY_VARS_NR_UE_s {
|
||||
uint8_t prs_active_gNBs;
|
||||
NR_DL_UE_HARQ_t dl_harq_processes[2][NR_MAX_DLSCH_HARQ_PROCESSES];
|
||||
NR_UL_UE_HARQ_t ul_harq_processes[NR_MAX_ULSCH_HARQ_PROCESSES];
|
||||
|
||||
//Paging parameters
|
||||
uint32_t IMSImod1024;
|
||||
uint32_t PF;
|
||||
@@ -540,6 +539,10 @@ typedef struct {
|
||||
int gNB_id;
|
||||
/// NR slot index within frame_tx [0 .. slots_per_frame - 1] to act upon for transmission
|
||||
int nr_slot_tx;
|
||||
/// NR slot index tx offset to resume
|
||||
/// in case of NTN, tx_offset can be changed dynamically via SIB19
|
||||
/// we need to notify the right tx thread slot based on TX offset change
|
||||
int nr_slot_tx_offset;
|
||||
int rx_slot_type;
|
||||
/// NR slot index within frame_rx [0 .. slots_per_frame - 1] to act upon for transmission
|
||||
int nr_slot_rx;
|
||||
|
||||
Reference in New Issue
Block a user