• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 99 subscribers
  • Views 1004 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Regarding Control info De multiplexing over PUSCH .

Naveen kumar2
Prodigy 130 points

Hi,

 I am running BCP Example Project for Uplink in which SSL and Rate Dematcher submodules of BCP are used for PUSCH symbol process. I am not able to find a way to demux HQ (Ack \Nack) control info from BCP  as the SSL theory says that it will overwrite the HQ bits with 0's and RI bits will be punctured. Can you please suggest what is the possible way to extract HQ data using BCP or we have to use the LTE Library functions for HQ demux as a separate thread or any other possibility if you can suggest ?

Thanks & Regards

Naveen Kumar

  • 0
    TI__Mastermind 30405 points

    Hi Naveen,

    BCP (SSL) punctures the RI bits and overwrites ACK/NACK softbits with 0 so you cannot get this information from BCP output. Control information needs to be extracted from the input data and processed seperately and TI's LTE library provides functions to handle this processing in software.

    Regards
    -Nitin

  • 0 in reply to Nitin Sakhuja
    Prodigy 130 points

    Hi Nitin,

    Thanks for the information. 

    But we have been wondering if BCP can locate the bits to be forced  to zero why not extract it. Is there no way BCP has information about HARQ bits?

    Our concern is running two threads when HARQ and CQI are present over ULSCH.

    Regards,

    Naveen

  • 0 in reply to Naveen kumar2
    TI__Mastermind 30405 points

    Hi Naveen,

    The BCP has not been designed to output control information along-with the PUSCH user data so the only way to get HARQ and CQI information is to extract the control symbols from input data in software, before sending it to the BCP.

    Regards
    -Nitin

  • 0 in reply to Nitin Sakhuja
    Prodigy 130 points

    Hi Nitin,

    Thanks for your answer. Yes you are correct that BCP does not support HI/RI over PUSCH demux but the sprugz clearly says that BCP can give CQI out data if CQI is present over PUSCH. I have verfied also that  whenever CQI is present over PUSCH BCP gives two packets , namely first packet for CQI data only and second for PUSCH data. But with this scenario I am facing a problem that CQI data is decoded properly but  CRC fails for PUSCH data. Is there any config that I am missing as a part of BCP header preparation. All CQI related Flags (CQI PassThrough, CQIPayload,CQI outbits) are taken care. Is there anything I need to take care regarding init_cb_flowId,flowId_hi,flowId_cqi_offset.  

    Thanks in advance.

    Thanks & Regards

    Naveen Kumar 

  • 0 in reply to Naveen kumar2
    TI__Mastermind 30405 points

    Hi Naveen,

    The BCP Rate dematcher does provide CQI output packet when the size of the CQI Payload is greater than 11. We'll need the following memory dumps in order to be able to look into this issue.

    1. BCP Header descritor
    2. BCP Header buffer
    3. BCP Payload descriptor
    4. BCP Payload buffer
    5. Expected CQI and PUSCH output data buffers.

    ***: In case you're using a single input packet for both header and payload, you can provide the input packet descriptor and input data buffer (containing both header and payload)

    Regards
    -Nitin

  • 0 in reply to Nitin Sakhuja
    Prodigy 20 points

    Hi Nitin,

    Got a tiny confusion about (O<=11 & O>11) conditions from your last post.

    -In case CQI Payload is smaller than 12, does process "skip" BCP and send data to RSA core?

    -Or does it still use BCP then is sended to RSA? Which processes are handled by BCP (O<=11).

    -If I can not get RSA support which is provided by 3rd party as mentioned earlier, I suppose I should implement a Fast Walsh Hadamard decoder in DSP. Is it possible in practice?

    Thanks a lot.

    Bedri