[srslte-users] Getting srsue to attach to PLMN

[Keith E. Fleming]

Greetings!
I am evaluating srsue for the purpose of acting as a UE in LTE Band 65
My hardware environment includes a Nuand BladeRF x40 and a Dell laptop (dual core, i5-3320M, 8GB). Started with a Raspberry Pi Model 4 but that turned out to be a lost cause.
I am using antennas that I have confirmed with my VNA they are appropriate for the UHF frequencies I am testing with. Reflection coefficient of -12 to -15db. So an SWR of approx 1.5. Not perfect, but acceptable.
I installed and compiled Nuand and ran the battery of tests (9,954 tests in total). I compiled version 21.04 of srsRAN. I ran cell_search on bands 2, 4, 12, 13, 14 and 66. I have a SIM in hand that I know the IMSI, IMEI, K and Opc. I put various EARFCN in ue.conf along with the SIM details and get wildly varying results.
Here is my attempt to attach to Verizon (EARFCN 5230, 751 MHz, -4.7dbm)

./cell_search -b 13
Found 2 cells
Found CELL 751.0 MHz, EARFCN=5230, PHYID=108, 50 PRB, 2 ports, PSS power=-6.8 dBm
Found CELL 751.0 MHz, EARFCN=5230, PHYID=47, 50 PRB, 2 ports, PSS power=-4.7 dBm

./srsueOpening 1 channels in RF device=default with args=default
Opening bladeRF...
Set RX sampling rate 1.92 Mhz, filter BW: 2.50 Mhz
Waiting PHY to initialize ... done!
Attaching UE...
Set RX sampling rate 1.92 Mhz, filter BW: 2.50 Mhz
set RX frequency to 751000000
set TX frequency to 782000000
.
Found Cell:  Mode=FDD, PCI=108, PRB=50, Ports=2, CFO=-0.4 KHz
Set RX sampling rate 11.52 Mhz, filter BW: 10.00 Mhz
Set RX sampling rate 1.92 Mhz, filter BW: 2.50 Mhz
.
Found Cell:  Mode=FDD, PCI=108, PRB=50, Ports=2, CFO=0.0 KHz
^CStopping ..
Couldn't stop after 5s. Forcing exit.
zsh: killed     ./srsue

I hit CTRL-C there above because it did *nothing*. No PCAP files were written to /tmp and the ue.log contents:

19:17:28.020766 [UE     ] [I] Using binary ./srsue with arguments:
19:17:28.020943 [UE     ] [I] Built in Release mode using commit 89f16eed2 on branch master.
19:17:29.016797 [PHY    ] [W] [    0] SYNC:  radio time seems to be going backwards (rx_time=0.026096, tti_ts=0.150577)
19:17:30.003953 [RRC    ] [W] FSM "selecting_cell" - Failed to select cell {pci=108, dl_earfcn=5230}
19:17:30.003968 [RRC    ] [E] Proc "Cell Search" - Couldn't select new serving cell
19:17:30.004719 [NAS    ] [W] Proc "NAS PLMN Search" - Did not find any PLMN in the set of frequencies.
19:17:30.008974 [PHY    ] [W] [    0] SYNC:  TTI jump of 5081 limited to 1000

I find it absurd that it couldn't find anything on that EARFCN. In the States, 751 MHz is Verizon's LTE backbone. I use it regularly to calibrate SDR clock offsets. It is a strong signal!
So.... Can anyone point me in the direction of getting this site to a stable running condition? There must be some parameters that need tweaking but my objective is to get this to work with a public ENodeB...
Thanks in advance!


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.srsran.com/pipermail/srsran-users/attachments/20210913/8c6c8412/attachment.htm>