[srslte-users] srsenb log and uplink mcs control

[J Giovatto]

Hi,

This might fix the log for you, duplicate phy_grant.n_prb[0] and missing
phy_grant.mcs.idx

--- a/srsenb/src/phy/phch_worker.cc
+++ b/srsenb/src/phy/phch_worker.cc
@@ -603,8 +603,8 @@ int phch_worker::decode_pusch(srslte_enb_ul_pusch_t
*grants, uint32_t nof_pusch)
       */
       log_h->info_hex(grants[i].data, phy_grant.mcs.tbs / 8,
                       "PUSCH: rnti=0x%x, prb=(%d,%d), tbs=%d, mcs=%d,
idx=%d, rv=%d, snr=%.1f dB, n_iter=%d, crc=%s%s%s%s%s%s%s%s\n",
-                      rnti, phy_grant.n_prb[0], phy_grant.n_prb[0],
phy_grant.L_prb,
-                      phy_grant.mcs.tbs / 8, phy_grant.mcs.idx,
grants[i].grant.rv_idx,
+                      rnti, phy_grant.n_prb[0], phy_grant.L_prb,
+                      phy_grant.mcs.tbs / 8, phy_grant.mcs.mod,
phy_grant.mcs.idx, grants[i].grant.rv_idx,
                       snr_db,
                       srslte_pusch_last_noi(&enb_ul.pusch),
                       crc_res ? "OK" : "KO",

Joe


On 06/04/2018 09:19 AM, Nicola Bui wrote:
> Dear SRS,
>
> I was testing srsENB with a commercial mobile phone trying to fix the
> MCS used in the communication, while I found a couple of problems:
> 1) the output of the PHY log of srsenb is not correctly aligned
> 2) setting pusch_mcs in the configuration file does not work
>
> Concerning the first I am reporting below three lines from the log
> file of the enb when I configured pusch_mcs = 5
> 11:30:54.070496 [MAC ] [I] [06649] SCHED: UL tx rnti=0x46, pid=1,
> dci=3,8, grant=(0,25), n_rtx=0, tbs=501, bsr=0 (512-501)
> 11:30:54.070515 [PHY1] [I] [06641] PDCCH: UL DCI Format0  rnti=0x46,
> cce_index=8, L=8, tpc=1, tti_tx_dl=6645
> 11:30:54.078372 [PHY1] [I] [06649] PUSCH: rnti=0x46, prb=(0,0),
> tbs=25, mcs=549, idx=10, rv=0, snr=31.3 dB, n_iter=1, crc=OK,
> dec_time=  67 us
> The first line is related to the MAC layer which schedules an uplink
> transmission for RNTI 0x46 using 25 resource blocks and a transport
> block size of 501 bytes
> The third line is the PHY layer log of the same transmission, which
> correctly report the RNTI, but, starting from tbs, all values are
> shifted of 1 position to the right (e.g., mcs cannot be 549,, but
> should be the next value, 10)
> From this last information I also understand that, even though I set
> pusch_mcs=5 the uplink transmission of the device is scheduled at MCS=10.
>
> I don't mind about the log misalignment, I can script a workaround to
> read the correct value, but I really like to test the mobile phone
> communication at fixed MCS on both uplink and downlink. Note that
> using pdsch_mcs correctly fix all downlink coommunications to the
> desired MCS. Is there anything I can do to fix this?
>
> Thanks,
> Nicola
>
>
> _______________________________________________
> srslte-users mailing list
> srslte-users at lists.softwareradiosystems.com
> http://www.softwareradiosystems.com/mailman/listinfo/srslte-users


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.srsran.com/pipermail/srsran-users/attachments/20180605/8d6b2bb6/attachment.htm>