- DM-RS type {1,2} was wrongly assigned to {1,2} instead of {0,1}
- Num DM-RS CDM groups without data is now taken into account in mthread
decoding function
- number of DM-RS CDM groups without data for format 1_0 according to 5.1.6.2
3GPP TS 38.214 version 15.9.0 Release 15:
-- the UE shall assume the number of DM-RS CDM groups without data is:
-- 1 (CDM group 0) for the case of PDSCH with allocation duration of 2 symb
-- 2 (CDM group {0,1}) for all other cases.
- moved RRC structs checks and mac configuration to config_ue.c
- new implementation of DCI configuration according to 3GPP TS 38.331
version 15.9.0 Release 15 and 3GPP TS 38.213 version 15.9.0 Release 15
- UE context status enters RRC_NR_NSA_RECONFIGURED state, after the eNB receives an RRCConnectionReconfigurationComplete message including the
scg_ConfigResponseNR_r15 container.
- Use this state to avoid triggering sGNB Addition Request multiple times
- replaced nr_slot_t with int as it was barely used in the code (also is defined after an lte struct)
- removed use of phy_frame_config_nr from the UE (another file was created in SCHED_NR_UE, still the
gNB file is used in some tests file at UE side)
- Msg2 scheduler used for testing purposes was removed
- Now the UE tries to DCI 1_0 scrambled with RA_RNTI in the RA window
- If no RAPID is found within the RA window, instructs PHY to TX PRACH
- IP data were fetched and passed to MAC directly at the PHY layer
- random data was generated whenever there's no upcoming data.
- Therefore these functionalities are now moved to MAC and FAPI interfaces
- are used instead.
- ceiling computation num of segments (cod/dec)
- fixes demodulation
- initialisation CW for PDSCH RA
- memcpy dl_ch_estimates to pdsch_vars_ra (hotfix)
- correct N_RB for DCI 1_0 and RNTI_RA
- LDPC decoder does not support lower MCS, therefore MCS 9 has been selected
for the Msg2 configuration.
- This is done only in do-ra mode, in order to prevent any fallout on the
current the configuration used for the operation with the commercial UE
hack: in schedule_ue_spec, 3 bytes are "reserved" (this should be
done better). An RLC AM entity may ask for only 2 bytes for
ACKing and get a TBS of 32 bits and due to these 3 reserved bytes we may
end up with only 1 byte left for RLC, which is not enough. This hack
prevents the problem. To be done better at some point. If the function
schedule_ue_spec is (has been) reworked, this hack can be removed.
Dig for "TBS - ta_len - header_length_total - sdu_length_total - 3"
in schedule_ue_spec.
This should be done better at some point.
cig