[srslte-users] Issue on SI / SIB messages

[Cedric Roux]

If you look at table 7.1.7.2.3-1 in 36.213
(which is the one to use for DCI format 1C for SI)
you see that TBS max is 1736.

For DCI format 1A we have in 36.212:
"- TPC command for PUCCH – 2 bits as defined in section 5.1.2.1 of [3]
    - If the format 1A CRC is scrambled by RA-RNTI, P-RNTI, or SI-RNTI:
      - The most significant bit of the TPC command is reserved.
                                                                   1A
      - The least significant bit of the TPC command indicates column N 
PRB of the TBS table defined of [3].
                                         1A             1A
       - If least significant bit is 0 then N PRB = 2 else N PRB = 3."

So you can only use 2 or 3 RBs, which means max TBS 2216
(table 7.1.7.2.1-1 of 36.213).

So I think it's the full message that must fit in this, not individual 
ones. You will need to schedule for SIBs in more than one message
I would say.

On 2021-03-01 16:00, Taillefer Simon wrote:
> Hello Cedric Roux,
> 
> We understood these lines as like a SIB (SIB1, SIB2, SIB7 and etc.) have a maximum size of 277 bytes but not the full message (called SI message) that contains all these SIBs. Please find attached an example of what I am trying to explain.
> 
> Our issue is that it is this full SI message that have the maximum size of 277 bytes and not the SIB content as we thought.
> 
> Thank you for your help,
> 
> Simon Taillefer
> 
> -----Message d'origine-----
> De : Cedric Roux <cedric.roux at eurecom.fr>
> Envoyé : lundi 1 mars 2021 13:56
> À : Taillefer Simon <simon.taillefer at telespazio.com>; srslte-users at lists.softwareradiosystems.com
> Objet : Re: [srslte-users] Issue on SI / SIB messages
> 
> see 3gpp document 36.331, 5.2.1.1, they say:
> "NOTE 1: The physical layer imposes a limit to the maximum size a SIB can take. When DCI format 1C is used the
>           maximum allowed by the physical layer is 1736 bits (217 bytes) while for format 1A the limit is 2216 bits
>           (277 bytes), see TS 36.212 [22] and TS 36.213 [23]. For BL UEs and UEs in CE, the maximum SIB and
>           SI message size is 936 bits, see TS 36.213 [23]. For NB-IoT, the maximum SIB and SI message size is
>           680 bits, see TS 36.213 [23]."
> 
> On 2021-02-22 18:23, Taillefer Simon wrote:
>> Hello srsLTE team,
>>
>> Basically, we are encountering an issue on the RRC protocol. We want
>> to send long SI / SIB messages but when the size is higher than 277
>> bytes, some errors are raised.
>>
>> More precisely, the errors are:
>>
>> -Error("Computing DL grant\n"); from
>> /srsLTE/srsenb/src/phy/cc_worker.cc
>> /file, around line 519 (depending on the srsLTE version and added code
>> on our side)
>>
>> -ERROR("Configuring TB Info\n"); from /srsLTE/lib/src/phy/phch/ra_dl.c
>> /file, around line 627.
>>
>> We found that these errors are coming from the
>> /grants[i].dci.tb[0].mcs_idx/ variable which has a value of -1
>> (4294967295exactly) instead of an unsigned integer. Then, we suppose
>> that the size of the corresponding transport block cannot be found in
>> the tbs_table (/srsLTE/lib/src/phy/phch/tbs_tables.h/file, line 38)
>> and this is why we have errors raised.
>>
>> So our questions are:
>>
>> -How this /grants[i].dci.tb[0].mcs_idx/variable is calculated?
>>
>> -Do you know why its value is -1 (4294967295) when the size of the SI
>> / SIB messages is longer than 277 bytes?
>>
>> -Maybe we are on the wrong way so do you know how it is possible to
>> send long SI / SIB messages?
>>
>> Thank you for your help,
>>
>> *Simon Taillefer***
>>
>>
>>
>>
>> -WARNING: This message contains confidential and/or proprietary
>> information which may be subject to privilege or immunity and which is
>> intended for the use of its addressee only.
>> Should you receive this message in error, you are kindly requested to
>> inform the sender and to definitively remove it from any paper or
>> electronic format.
>> Any other use of this e-mail is strictly forbidden. Thank you in
>> advance for your cooperation.
>>
>> Please consider the environment before printing this e-mail.
>>
>> _______________________________________________
>> srslte-users mailing list
>> srslte-users at lists.softwareradiosystems.com
>> http://www.softwareradiosystems.com/mailman/listinfo/srslte-users
>>
> 
> 
> 
> -WARNING: This message contains confidential and/or proprietary information which may be subject to privilege or immunity and which is intended for the use of its addressee only.
> Should you receive this message in error, you are kindly requested to inform the sender and to definitively remove it from any paper or electronic format.
> Any other use of this e-mail is strictly forbidden. Thank you in advance for your cooperation.
> 
> Please consider the environment before printing this e-mail.
>