GitLab

Necessary modifications to support noS1 with nfapi and eNB side for testing vnf-pnf up to the stage of previous commit

Testing works until the call of callback function pnf_p7_subframe_ind. There, it never satisfies the condition if(tx_subframe_buffer->sfn_sf == sfn_sf_tx) so both sides (vnf and pnf are hanging).

Connected new callback functions at nfapi_pnf for phy_stub mode + Intorducing new functions initializing UE variables and threads specific to phy_stub mode, instead of extending the existing ones + Direct connection of the timer_thread with UE_phy_stub_thread_rxn_txnp4.

Introduced a new timer_thread within lte-ue. This thread is ticking the subframes and wakes up UE_thread function which is waiting on a condition variable for nfapi_mode=3 (phy_stub)

Introduce nfapi_mode variable to differentiate functionality between phy_stub_ue mode (nfapi_mode=3) and normal full stack mode (nfapi_mode!=3).

Added functions copying nfapi requests to UE_mac_inst + Substituted calls to phy_procedures_ue rx/tx with phy_stub functions triggering MAC layer Tx/Rx procedures at UE MAC. Changes within UE_thread_rxn_txnp4

uplink data from ue received, corrected scheduling of DCI0/UPLINK_CFG.  Fails to correctly form (SFNSF) for PHICH ACK

Change +4 for tx to a variable and get it to run at +2, this alllows nFAPI to get the reply back for HI_DCI0 to generate the PHICH in time

Crashes around msg4 time, but understand that need to get the ACK for the UE UL of msg3 back to UE within 4ms

basically the T_xx variables are int, not uint

phy_init_lte_ue_signal was renamed init_lte_ue_signal in openair1/PHY/INIT/defs.h
maybe it's the opposite that should have been done?

Functions used in one file and defined in another may have been
declared in the wrong header file.

modifications for TDD and IF4p5 (in particular TDD). Both in RRU and RAU components. Monolithic is unchanged.

added LUT for finer-grain ULSCH power measurement compensation based on TBS. Still seems too high, to be checked some more.

see commit 622b919b

removed "short_offset" code from normal_prefix_mod. removed static declaration in PHY_ofdm_mod which was a cause for problems when parallelizing FFTs for FEP TX.

Merge from RU-RAU-split and fix of frame/subframe handling.  Was very confused with the counter in the main loop, the incrementing from the USRP read library and the nFAPI frame handling.  RACH seen on wireshark and something happens on VNF