- 03 May, 2021 1 commit
-
-
Sakthivel Velumani authored
The CP length was hard coded for numerology 1
-
- 30 Apr, 2021 2 commits
-
-
Sakthivel Velumani authored
-
Sakthivel Velumani authored
-
- 29 Apr, 2021 1 commit
-
-
Sakthivel Velumani authored
adding option to change numberology
-
- 27 Apr, 2021 3 commits
-
-
Thomas Schlichter authored
-
Thomas Schlichter authored
-
Remi Hardy authored
MR !1079 / MR !1108: 1. Read slots from config file and schedules all slots *except* for slots 0 and 10. This is not a problem, since currently we cannot allocate more than six HARQ feedbacks in a slot, but we might have up to 7 with current configuration 2. phytest mode is configurable. From `nr-softmodem -h`: -m: Set the downlink MCS for PHYTEST mode -t: Set the uplink MCS for PHYTEST mode -M: Set the downlink banwdwidth (in PRBs) for PHYTEST mode -T: Set the uplink banwdwidth (in PRBs) for PHYTEST mode -D: Bitmap for DLSCH slots (slot 0 starts at LSB) -U: Bitmap for ULSCH slots (slot 0 starts at LSB) 3. Scheduler prints more information: UE PHR, PUCCH SNR, PUSCH SNR, RSSI 4. The scheduler learned to scheduler DLSCH/ULSCH in mixed slots (if the time domain allocation matches the mixed slot) 5. The scheduler expects * PDSCH Time Domain allocation index 0 to be the DL slot allocation * PDSCH TDA index 1 for Mixed DL slot * PUSCH TDA index 0 for UL slot * PUSCH TDA index 1 for Mixed UL slot * PUSCH TDA index 2 for Msg.3 6. Some more cleanup, e.g., separate PDSCH/PUSCH data structures into "semi-static" data (e.g., TDA allocation to use, since it changes seldomly) and "dynamic" data (e.g., RB allocation per slot, which changes every slot) MR !1137 : Benetel config files fix
-
- 26 Apr, 2021 10 commits
-
-
Sakthivel Velumani authored
-
Thomas Schlichter authored
The TA calculation takes the CIR peak position error and scales it according to 38.213 section 4.2. Till now it was simply an integer division "timing_advance_update = sync_pos / bw_scaling". E.g. in case of bandwidth_scaling == 16, the timing_advance_update was 0 for sync_pos between -15 and +15. With this change, this range (and thus the UL timing error) will be reduced to -8 and +8.
-
Thomas Schlichter authored
Only Ncp1 is used, so get rid of the second timeshift_symbol_rotation array. Also simplify the computation of "t0 * Tc * f".
-
Thomas Schlichter authored
The rotation compensating the shifted OFDM position _before_ the channel estimation makes the CIR peak move back to 0. We have to take this into account on gNB side when calculating the timing advance command, and on UE side in the time adjustment.
-
Thomas Schlichter authored
-
Sakthivel Velumani authored
-
Sakthivel Velumani authored
-
hardy authored
-
Sakthivel Velumani authored
-
Sakthivel Velumani authored
for gNB its in L1 section of config file for UE its a cmd line paramenter
-
- 25 Apr, 2021 2 commits
-
-
Robert Schmidt authored
-
Robert Schmidt authored
-
- 24 Apr, 2021 1 commit
-
-
Remi Hardy authored
MR !1129 : gnb-realtime-hotfix -Hotfix for realtime performance issue -Enabled L1 and scheduler timing statistics -Changed order of UL Indication. L1 Rx -> UL Ind -> L1 Tx MR !1123 : [CI] ci_phytest -new 5G NR phy test -q -U 787200 -T 106 -t 28 -D 130175 -m 28 -M 106 MR !1128 : [CI] ci_add_runtime_stats -adding L1 processing stats MR !1132 : [CI] ci_add_uldlharq_stats -additonal ulsch/dlsch stat no MR : hotfix branch fixgtpu -revert default values for 5GS -fix T_IDs.h build
-
- 23 Apr, 2021 4 commits
-
-
Sakthivel Velumani authored
also fixed build wanings
-
Sakthivel Velumani authored
This reverts commit 87b53883.
-
Sakthivel Velumani authored
This is necessary to compensate the rotation caused by the timeshifted OFDM symbol (1/8 CP)
-
Remi Hardy authored
-
- 22 Apr, 2021 3 commits
- 21 Apr, 2021 13 commits
-
-
hardy authored
-
Laurent THOMAS authored
-
Laurent THOMAS authored
-
Thomas Schlichter authored
Currently we cute the PUSCH and PDSCH OFDM symbols starting within the last 1/8th of the CP to avoid ISI. This time shift leads to a rotation in frequency domain. The current PRB based channel estimation does not compensate for the rotation _within_ a PRB, leading to bad performance. This commit should _not_ make it into develop branch, it is just for testing! Instead, we should do some compensating rotation before channel estimation.
-
Thomas Schlichter authored
-
Thomas Schlichter authored
-
Thomas Schlichter authored
use nb_antennas_tx instead of nb_antennas_rx when copying data from UE->common_vars.txdata to s_re and s_im
-
Thomas Schlichter authored
Each element is sizeof(int32_t), not sizeof(int32_t *).
-
Florian Kaltenberger authored
reverting some changes in benetel-5g config files that were introduced in anticipation of MR 1079, but which has not been merged yet.
-
Laurent THOMAS authored
-
Laurent THOMAS authored
-
Laurent THOMAS authored
-
hardy authored
-