Commit Graph
1335 Commits
Author SHA1 Message Date
Cedric Roux 668d7af7d2 align variables on stack in rank_estimation_tm3_tm4
dlsim_tm4 was crashing on massive. It was a problem
of misalignment of variables. The generated code by
the compiler was using 'movdqa' which requires alignment
to 16 bytes. Let's put 32, just in case, for avx2 maybe
(maybe not necessary but should not hurt).
2017-03-21 14:12:00 +01:00
Elena_Lukashova 79c304a24a Passing Conditional Number threshold as a command line argument. -V($) 2017-03-21 12:13:17 +01:00
Elena_Lukashova ca0d8a445e Fixing dlsch_demod_shift. 2017-03-21 12:03:23 +01:00
Elena_Lukashova 93c1315bbd Adequate throughput computation for rank adaptation. 2017-03-21 02:13:08 +01:00
Elena_Lukashova b596ede727 Merge remote-tracking branch 'origin/develop' into feature-59-tm4 2017-03-20 14:33:03 +01:00
Cedric Roux e2350103f4 hotfix: LLR was calling qam64_qam16_avx2 instead of qam64_qam64_avx2
Not sure of this one, to be checked by TCL.
2017-03-20 13:33:27 +01:00
Cedric Roux 1218cc8f65 bugfix: fix dump_dlsch2
the argument 'coded_bits_per_codeword' has to be an array in
case of several codewords.

The calling sites have been adapted.

Today, only the first index is used, so calling sites where
'coded_bits_per_codeword' is an integer pass the address
of it. It is expected that 'dump_dlsch2' will check in the
future that there is one or two codewords and only access
'coded_bits_per_codeword[1]' when it's sure there are
really two codewords.
2017-03-20 11:15:57 +01:00
Cedric Roux 7ffc20426b integration fix: let simulators compile again
- pucchsim
- prachsim
- pdcchsim
- pbchsim
- mbmssim

Mostly, add and initialize the variable 'cpuf'.
Also, pdcchsim was calling rx_pdcch with wrong
arguments.
2017-03-20 11:00:41 +01:00
Cedric Roux 5014b84c30 integration fix: let ulsim compile again + minor modification in dlsim_tm4 2017-03-20 10:43:50 +01:00
Cedric Roux f2e9e3f2dc integration fix: let dlsim compile again
Same as previous commit 41dda3d3de
but for dlsim.
2017-03-20 10:40:26 +01:00
Cedric Roux 41dda3d3de integration fix: let dlsim_tm4 compile again
The variables that were duplicated per processing thread
were adapted. 'cpuf' was defined and initialized.
2017-03-20 10:38:25 +01:00
gabrielC 1e2f952a9c Remove to frequent log in 20 MHz 2017-03-20 09:18:02 +01:00
Elena_Lukashova b835f63956 1. There is probably a bug in generate_ue_dlsch_from_dci
for DCI format 2. Temporaryly going back to the previous
version of code.
2. Enabling rate adaptation with multiple HARQ rounds.
(no change inside the rounds).
2017-03-20 01:06:12 +01:00
Elena_Lukashova a92ed31914 Merge branch 'develop' into feature-59-tm4 2017-03-19 15:52:45 +01:00
Cedric Roux 2011c7d689 Merge remote-tracking branch 'origin/ulsim-bugfix-w11' into develop_integration_w11 2017-03-18 00:10:50 +01:00
Cedric Roux 1c57eadd33 fix: remove useless test
One check introduced by previous commit was not necessary.
2017-03-18 00:09:48 +01:00
Cedric Roux 7847143610 Merge remote-tracking branch 'origin/bugfix-224-ue-tcrnti' into develop_integration_w11 2017-03-17 19:33:56 +01:00
fnabet 7f9b19cce8 UE Issue224 fix on TCRNTI 2017-03-17 10:52:59 +01:00
Florian Kaltenberger c13d5c21f3 added null pointer check for ulsim 2017-03-16 18:05:23 +01:00
Cedric Roux 639d2eefc6 Merge remote-tracking branch 'origin/phy-test-fix' into develop_integration_w11 2017-03-16 18:01:23 +01:00
gabrielC d5a9334306 20MHz improvement for scope (still need to be fixed for 20MHz LLR) and Log 2017-03-16 17:06:34 +01:00
gabrielC 950c1c060e Correction of dlsch channel calculation 2017-03-16 11:45:07 +01:00
Cedric Roux f5aa0135b1 hotfix: fix buffer size for phy-test mode 20MHz
768*8 is too small, in 20MHz, the phy-test mode
(in default mode?) has a buffer of size 6378.

Let's set the buffer size to the input's one plus 4 bytes
(to accomodate for CRC, not sure about this).
2017-03-15 17:17:54 +01:00
gabrielC 752b305270 Merge branch 'pdcch_dual_buffer' into develop_integration_w11 2017-03-14 15:44:13 +01:00
gabrielC 28a50c15e9 Remove some print and correct rnti for the pdcch_vars odd/even subframe 2017-03-14 15:35:21 +01:00
Elena_Lukashova 488933eddc Removing unnecessary minus_one and plus_one. 2017-03-13 17:04:36 +01:00
hbilel bd4898222f pdcch dual buffer 2017-03-13 15:50:10 +01:00
Cedric Roux a500b2b293 integration fix: fix compilation of ulsim
We use the odd/even dlsch corresponding to the configured
subframe. It may not be the right thing to do.
2017-03-13 15:43:43 +01:00
Elena_Lukashova 21481c4a6a Adding check for the first transmission in prepare_dl_decoding_format1_1A. 2017-03-13 15:13:45 +01:00
hbilel cec9cb7662 fix compilation error for dlsim 2017-03-13 12:05:47 +01:00
hbilel b82b7781ee integration fix clear tx-buffer of ue in case of empty subframe
for all processing including oaisim
2017-03-13 10:14:26 +01:00
Cedric Roux a194009e00 integration fix: put dlsch_demod_shift into dlsch_demodulation 2017-03-13 10:06:26 +01:00
Elena_Lukashova e5908bfdaa Fixing a bug in conjch0_mult_ch1 2017-03-12 00:29:50 +01:00
Elena_Lukashova 14d927d414 Fixing TM condition for rank estimation.
the eNB_id was overwriiten by the preceding loop
and setting ue->transmission_mode = 0.
2017-03-11 19:25:15 +01:00
Elena_Lukashova 263b7aa07a Fixing typo. 2017-03-11 18:48:57 +01:00
Elena_Lukashova 9fd0f35134 In rank estimation rank 1 is for CLSM and OLSM, 0 is for Tx div. 2017-03-10 16:45:44 +01:00
Elena_Lukashova 83a045ec83 Condition to compute Rank only for TM3, TM4. 2017-03-10 16:37:17 +01:00
Elena_Lukashova e709933695 Fixing printouts. 2017-03-10 16:36:07 +01:00
Elena_Lukashova 132018a136 Adding Rank estimation for TM4.
Right now no feedback, only calculation.
To see the values, enable DEBUG_RANK_EST
in lte_ue_measurements.
2017-03-10 16:20:41 +01:00
Elena_Lukashova e704a3155c Enabling noise in the whole subframe and computing N0_total. 2017-03-10 16:16:54 +01:00
gabrielC 4abc1d21e3 20 MHz UE fix 2017-03-09 16:33:07 +01:00
gabrielC 28c009a072 Chenge LOG_I to LOG_D 2017-03-09 10:20:52 +01:00
hbilel 305ebf5c4b rebase branch to develop 2017.w09 2017-03-08 14:57:30 +01:00
hbilel 9dd65f2801 [OAI-UE] remove unused debug-printf 2017-03-08 14:21:14 +01:00
hbilel 05fe029a62 [OAI-UE] fix in de-precoding for tm3/4 2017-03-08 13:59:49 +01:00
hbilel 3d1e01e283 [OAI-UE] llr computation clean up 2017-03-08 12:16:38 +01:00
hbilel bb8c6b2c52 [OAI-UE] code clean up for merge 2017-03-08 12:16:38 +01:00
hbilel 3dc1963a2d [OAI-UE] MIMO fixes + disabling MIB/SIB decoding on connected mode 2017-03-08 12:16:38 +01:00
Sebastian Wagner 42b834d64c Changes for 64qam_64qam for avx2
Conflicts:
	openair1/PHY/LTE_TRANSPORT/dlsch_llr_computation.c
2017-03-08 12:16:38 +01:00
hbilel ba55fcac27 [OAI-UE] Mimo LLR computation AVX2 + fix in TM3 deprecoding 2017-03-08 12:16:38 +01:00