[srslte-users] Getting srsUE up and running

Wed Jul 27 12:44:47 UTC 2016

I'm trying with 10 MHz BW....but still, MIB decoding fails most of the time and even if it succeeds then it fails to sync :(. 

> From: "Ismael Gomez" <ismael.gomez at softwareradiosystems.com>
> To: "Tomcsányi," <domi at tomcsanyi.net>
> Cc: "srslte-users" <srslte-users at lists.softwareradiosystems.com>
> Sent: Wednesday, July 27, 2016 2:24:48 PM
> Subject: Re: [srslte-users] Getting srsUE up and running

> Exactly. With a 20 MHz signal is nearly impossible the the ue can proceed with
> the decoding process. Can you try with a quadcore or reducing the bw to 10 MHz?
> On Wed, 27 Jul 2016, 14:13 Tomcsányi,, < domi at tomcsanyi.net > wrote:

>> Yes, I understand this, it makes sense. My only issue is that I can't get srsUE
>> to decode MIB correctly, and also if once it finally detects the correct BW it
>> fails to synchronize - with all frequency scaling changed to maximum
>> performance.
>> Is there any way to hard-code MIB data so at least that part is taken care of?
>> Or is this a bad idea, because if MIB decoding fails it means the radio
>> connection is such low quality between the eNB and the UE that there is no way
>> it's going to work?

>> Cheers,
>> Domi

>>> From: "Ismael Gomez" < ismael.gomez at softwareradiosystems.com >
>>> To: "Tomcsányi," < domi at tomcsanyi.net >
>>> Cc: "srslte-users" < srslte-users at lists.softwareradiosystems.com >

>>> Sent: Wednesday, July 27, 2016 1:58:58 PM
>>> Subject: Re: [srslte-users] Getting srsUE up and running

>>> As I said before, in order to synchronize and decode the 20 MHz signal you'll
>>> need a true 4-core system (your processor is 2-core). Notice that generating a
>>> signal (what the eNodeB does) requires much less processing than decoding it
>>> (what the UE does), this is why with the eNodeB it is possible to generate the
>>> signal with the same laptop (probably not support high UL rates).

>>> On Wed, 27 Jul 2016 at 13:52 Tomcsányi, < domi at tomcsanyi.net > wrote:

>>>> Yes, it ran in 20 MHz mode, but I didn't do much of actual IP traffic, just an
>>>> attach.
>>>> So I rebooted my PC with SpeedStep and C states disabled, currently I'm having a
>>>> hard time decoding the MIB. It still detects random BW values (right now it
>>>> detected after many tries 1.4 MHz which is wrong again). I don't know what to
>>>> do, will keep trying to adjust the gain settings maybe I can find a good value
>>>> finally.

>>>> Cheers,
>>>> Domi

>>>>> From: "Ismael Gomez" < ismael.gomez at softwareradiosystems.com >
>>>>> To: "Tomcsányi," < domi at tomcsanyi.net >

>>>>> Cc: "srslte-users" < srslte-users at lists.softwareradiosystems.com >
>>>>> Sent: Wednesday, July 27, 2016 1:36:56 PM
>>>>> Subject: Re: [srslte-users] Getting srsUE up and running

>>>>> Hi Domi,
>>>>> were you able to run OAI eNodeB in 20 MHz mode? With that processor, srsUE
>>>>> should do full rate downlink in 10 MHz mode and support up to 35 Mbps. For 20
>>>>> MHz bw and to reach 75 Mbps downlink you'll need a true 4-core processor.

>>>>> On Wed, 27 Jul 2016 at 13:33 Tomcsányi, < domi at tomcsanyi.net > wrote:

>>>>>> Hi Ismael,

>>>>>> Thanks for the quick reply. I'm using an i7-6500U CPU @ 2.5GHz. I'll try to
>>>>>> disable frequency scaling and also set it to maximum performance, however I
>>>>>> find it weird that I was able to run a full eNodeB with OpenAirInterface5G
>>>>>> using the same hardware (laptop and B200) with the same frequency scaling
>>>>>> settings and phones were able to connect to it fine.
>>>>>> Also I asked my colleague who configured the eNB and it is set for 10 MHz
>>>>>> bandwidth.

>>>>>> I'll reboot now and disable frequency scaling in BIOS/EFI.

>>>>>> Cheers,
>>>>>> Domi

>>>>>>> From: "Ismael Gomez" < ismael.gomez at softwareradiosystems.com >
>>>>>>> To: "Tomcsányi," < domi at tomcsanyi.net >, "srslte-users" <
>>>>>>> srslte-users at lists.softwareradiosystems.com >
>>>>>>> Sent: Wednesday, July 27, 2016 1:26:01 PM
>>>>>>> Subject: Re: [srslte-users] Getting srsUE up and running

>>>>>>> Hi Domi,
>>>>>>> It looks like it's a CPU issue. Make sure the CPU frequency scaling is disabled
>>>>>>> and the frequency governor is set to "performance" (ie fixed to maximum MHz).
>>>>>>> Also make sure you don't run other background processes. What sort of CPU are
>>>>>>> you using?

>>>>>>> Do you have access to the enodeb to try with a lower BW, like 10 MHz?

>>>>>>> On Wed, 27 Jul 2016 at 13:21 Tomcsányi, < domi at tomcsanyi.net > wrote:

>>>>>>>> Hi all,

>>>>>>>> I'm sorry if this has been asked before on the list, but I couldn't find any
>>>>>>>> info about this using Google, so I'm turning to you:
>>>>>>>> I'm trying to connect srsUE to an LTE test network consisting of a commercial
>>>>>>>> eNB and MME. All my phones can see the network, connect to it etc. but with
>>>>>>>> srsUE I see the following:

>>>>>>>> root at LTE:~/srs/srsUE/build/ue/src# ./ue ue.conf
>>>>>>>> linux; GNU C++ version 4.8.4; Boost_105400; UHD_003.009.003-release

>>>>>>>> --- Software Radio Systems LTE UE ---

>>>>>>>> Reading configuration file ue.conf...
>>>>>>>> Using srsLTE version 001.002.000
>>>>>>>> Opening USRP with args: type=b200,recv_frame_size=9232,send_frame_size=9232
>>>>>>>> -- Detected Device: B200
>>>>>>>> -- Operating over USB 3.
>>>>>>>> -- Detecting internal GPSDO.... Found an internal GPSDO
>>>>>>>> -- Initialize CODEC control...
>>>>>>>> -- Initialize Radio control...
>>>>>>>> -- Performing register loopback test... pass
>>>>>>>> -- Performing CODEC loopback test... pass
>>>>>>>> -- Asking for clock rate 16.000000 MHz...
>>>>>>>> -- Actually got clock rate 16.000000 MHz.
>>>>>>>> -- Performing timer loopback test... pass
>>>>>>>> -- Setting master clock rate selection to 'automatic'.
>>>>>>>> -- Asking for clock rate 32.000000 MHz...
>>>>>>>> -- Actually got clock rate 32.000000 MHz.
>>>>>>>> -- Performing timer loopback test... pass
>>>>>>>> Setting frequency: DL=2685.0 Mhz, UL=2565.0 MHz
>>>>>>>> Searching for cell...
>>>>>>>> Found CELL ID: 1 CP: Normal , CFO: 1.8 KHz.
>>>>>>>> Trying to decode MIB...
>>>>>>>> - Cell ID: 1
>>>>>>>> - Nof ports: 2
>>>>>>>> - CP: Normal
>>>>>>>> - PRB: 100
>>>>>>>> - PHICH Length: Normal
>>>>>>>> - PHICH Resources: 1/6
>>>>>>>> - SFN: 0
>>>>>>>> MIB received BW=20 MHz
>>>>>>>> Initializating cell configuration...
>>>>>>>> Setting Sampling frequency 23.04 MHz
>>>>>>>> No space for SSS/CP detection. Realigning frame...
>>>>>>>> No space for SSS/CP detection. Realigning frame...
>>>>>>>> No space for SSS/CP detection. Realigning frame...
>>>>>>>> Timeout while synchronizing SFN
>>>>>>>> No space for SSS/CP detection. Realigning frame...
>>>>>>>> No space for SSS/CP detection. Realigning frame...
>>>>>>>> Timeout while synchronizing SFN
>>>>>>>> No space for SSS/CP detection. Realigning frame...
>>>>>>>> No space for SSS/CP detection. Realigning frame...
>>>>>>>> Timeout while synchronizing SFN
>>>>>>>> No space for SSS/CP detection. Realigning frame...
>>>>>>>> Timeout while synchronizing SFN
>>>>>>>> No space for SSS/CP detection. Realigning frame...
>>>>>>>> No space for SSS/CP detection. Realigning frame...
>>>>>>>> No space for SSS/CP detection. Realigning frame...
>>>>>>>> Timeout while synchronizing SFN
>>>>>>>> Timeout while synchronizing SFN
>>>>>>>> No space for SSS/CP detection. Realigning frame...
>>>>>>>> No space for SSS/CP detection. Realigning frame...

>>>>>>>> I'm using an USRP B200 with GPSDO (so not the MIMO capable B210 version) as you
>>>>>>>> can see. I'm pretty sure my ue.conf is in tact, basically just modified the
>>>>>>>> frequency and the USIM parts of it. I tried to mess around with the gain
>>>>>>>> settings since the whole setup is on my table, so not much gain is needed in my
>>>>>>>> opinion, but that didn't change much. It is weird though that MIB decoding
>>>>>>>> gives different results each time I start srsUE. The attached log you can see
>>>>>>>> shows 20 MHz of BW, but it also detects 10 MHz regularly.
>>>>>>>> Also sometimes it just says "Error receiveing from UHD: 30" and then I need to
>>>>>>>> unplug and plug in the USRP to get it working again.

>>>>>>>> I hope you can shed some on this for a newbie like me.

>>>>>>>> Cheers,
>>>>>>>> Domi
>>>>>>>> _______________________________________________
>>>>>>>> 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/20160727/5f775b23/attachment.htm>