[srslte-users] Some commercial UEs can't detect the enb

[Yi Chu]

Hi Paul,

I have tested the B205mini with the external 10MHz reference from an
Octoclock-G (GPSDO enabled), there are some strange logs in the enb but the
UEs can now find and attach to the enb! Please see the logs below:

---  Software Radio Systems LTE eNodeB  ---

Reading configuration file /home/yc541/.config/srslte/enb.conf...
Opening 1 RF devices with 1 RF channels...
[INFO] [UHD] linux; GNU C++ version 7.5.0; Boost_106501; UHD
_4.0.0.0-531-g5fb585c3
[INFO] [LOGGING] Fastpath logging disabled at runtime.
Opening USRP with args: "",master_clock_rate=23.04e6
[INFO] [B200] Detected Device: B205mini
[INFO] [B200] Operating over USB 3.
[INFO] [B200] Initialize CODEC control...
[INFO] [B200] Initialize Radio control...
[INFO] [B200] Performing register loopback test...
[INFO] [B200] Register loopback test passed
[INFO] [B200] Asking for clock rate 23.040000 MHz...
[INFO] [B200] Actually got clock rate 23.040000 MHz.
Could not lock reference clock source. Sensor: ref_locked=false, UHD error:
0
Setting frequency: DL=2650.0 Mhz, UL=2530.0 MHz
Setting Sampling frequency 5.76 MHz

I'm not sure what the red bit means but the 10MHz reference should be
working. The B205mini only has one ref port, according to the Ettus website
it can take either 10MHz or 1pps. Maybe that's the reason? Anyway my UEs
are working perfectly, many thanks for the help!

Regards
Yi

On Thu, 6 Aug 2020 at 11:44, Yi Chu <yi.chu at york.ac.uk> wrote:

> Hi Paul,
>
> Many thanks for getting back!
>
>>
>> 2) Check that there are no interfering signals on the frequencies of use
>>     - B1 and B7 are fully allocated in the UK (
>> https://www.spectrummonitoring.com/frequencies/#UK) so use a Faraday
>> cage or check for existing carriers
>>
>
> I used the USRP as a spectrum analyser and checked the LTE bands in my
> area, the frequencies I picked are not occupied at the moment. I might be
> able to run the tests in an anechoic chamber at some point.
>
>
>> 3) use a GPSDO or external clock source, or make sure carrier frequency
>> offset is small enough
>>     - Are you using an external clock source or do you have access to
>> equipment to check the frequency accuracy?
>>
>
> I have an external clock source with GPSDO but I don't have access to it
> immediately, I will try that once I can.
>
>
>> 4) Enable logs on the eNodeB and check if PRACH is detected
>>
>
> I enabled the debug level enb logs, but the last logs I can see is
> normally the S1 setup response.
>
>
>> 5) Try adjusting tx_gain, reducing it or increasing. If the UE sees low
>> power (<pmax threshold), it won't PRACH
>>
>
> I have tried tx_gain from 80 to 120, the UE is within 2 meters from the
> antenna. How is the gain mapped to the actual gain of the USRP? I can't
> find it in the enb.log or the logs shown in the terminal.
>
> For the iPhone 7 case, you could compare PRACH, SIB1 and SIB2
>> configurations for differences between srsENB and OAI. You could also add
>> SIB3 configuration in case the UEs require it.
>>
>
> I will try matching the SIB1 and SIB2 configurations and give the iPhone
> 7 another go.
>
>
>> Are you using the default config files for everything except EARFCN? If
>> not, could you share your config files?
>>
>
> I only modified the enb and rf sections, anyway here is the enb.conf.
>
> The other minor thing to mention is that I'm using the 19_12 release which
> is stable for VoLTE setup. I have tested the 20_04 release with the Huawei,
> iPhone, Samsung S5 and MI Mix 2 but the results are the same.
>
> Regards
> Yi
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.srsran.com/pipermail/srsran-users/attachments/20200818/0200e1b6/attachment.htm>