• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 101 subscribers
  • Views 712 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TCI6630K2L: Issue with SSL module in BCP for LTE Uplink

Ashok H.S.
Expert 1610 points

Part Number: TCI6630K2L

Hi All,

I am using the BCP for for LTE Uplink. The test project given in the "pdk_keystone2_3_01_03_06" works fine for UL Test. Now I want to test it with my own input.

As a 1st step I am testing only SSL output.

I have created a test case with almost same parameters (Cell Id, RNTI, TB size, number of RB,  modulation type etc). Tapped the output of the IDFT module on the reference C code what I have. And fed it to BCP. I have taken care of the partial de-interleave requirement of the SSL input here. I have verified it with hard coded indices in the input buffer instead of the actual IQ values for each symbol. 

1. As a 1st step I have disabled descrambler also. Then when I check the SSL output I see that it has a sign reversal when compared my reference C code output for the same. i.e., when the reference output is +ve, SSL output is -ve and vice versa. Is there anything missing here..? Or is it expected..? Is there any setting to be changed to rectify this..?

2. Assuming this is expected (sign reversal), I enabled the descrambler for the same. Now the descrambled soft bits not at all matching in signs whether in normal mode or in sign reversal mode.

What am I missing here..?

I have used a SNR of 1 and a Q format of 0, and an RMS value of 1 hence my SSL output is +/- 127 for QPSK.

Please let me know if you need any other info.

Regards,

Ashok

  • 0
    TI__Guru 65405 points
    Hi Ashok,

    I've forwarded this to the experts. Their feedback should be posted here.

    BR
    Tsvetolin Shulev
  • 0 in reply to Cvetolin Shulev-XID
    Expert 1610 points

    Thanks Tsvetolin Shulev,

    It would be great if some one clarifies how the descrambler works on BCP.

    In our reference code we have used the following equation to do the descrambling:

    DeScramBits[i] = SoftBits[i] * (1-2*ScramSeq[i]);

    I have made sure that the Cinit is matching both on the reference and test codes. So I guess scrambling seq as such should not have any issue.

    So in effect if the Scrambling Seq bit is 0, then the corresponding softBit's sign is unchanged and if its 1, then the soft bit's sign is reversed.

    If BCP is doing the descrambling in any other way then I may be interpreting the result wrongly...

    Thanks & regards,
    Ashok

  • 0 in reply to Ashok H.S.
    Expert 1610 points
    Hi All...

    I could solve the issue...!!

    My input to the SSL module was found to be wrong.

    I had prepared the input as per the BCP document (section 4.7.1.1, figure 4-73) and the output of SSL never matched.

    Finally I found a code for this partial deinterleaving on one of the other posts in the forum

    e2e.ti.com/.../223880

    Which worked...!! But the out of this partial deinterleaver and the description in the BCP document are totally different..Not sure why...

    Regards,
    Ashok