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.

VCP2 tailbiting decoding for TCI6614

Other Parts Discussed in Thread: TMS320TCI6614

I am trying to use the VCP2 example code which I found in forum for decoding tailibiting convolutionally encoded data. The issue I am facing is that output (viterbi output/decoded data) is not matching with expected output. I tried different  methods suggested by our friends in forum but the output is not matching. Here are the parameters I am giving as input to the VCP2:

Rate:1/3, Constraint length:7, length:32 , tailBitEnable:1,Decisions: Hard, poly0: 0x5B, poly1:0x79, poly2:0x75

Input bits: 01010101010101010101010101010101

encoded streams: 0xAAAAAAAA (1st), 0xAAAAAAAA (2nd), 0x55555555 (3rd)

Modulation: QPSK

EVM part number: TMS320TCI6614

Can someone help me in this?

Thanks & Regards,

Anjibabu

  • Hi,
    We are working on your post. Thank you for your patience.
  • Hi Anjibabu,

    Your polynomial register configuration seems to be incorrect. The correct polynomial for 1/3 rate is:

    VcpConfigParms.poly0 = 0x6C;

    VcpConfigParms.poly1 = 0xE4;

    VcpConfigParms.poly2 = 0xD4;

    VcpConfigParms.poly3 = 0x00;

    Please refer to the following thread for more details.

    -Nitin

  • Hi Nithin,

    Thank you for reply. I have changed polynomials and tried the same way in the link that you have provided but still not getting correct output. Can you please help me in this, it's been long that I struck with this issue.

    Thanks & Regards,

    Anjibabu

  • Have you tried the original example as provided to make sure that it is working for you before changing the parameters?

    -Nitin

  • Nithin,

    Yes,  I tried the example code before changing the parameters .  It is working fine with convolutinal encoder. But if  I use  tailbiting convolutinal encoding it is not working (not giving desired output).

    Regards,

    Anjibabu

  • How many bits from the beginning of the input are you copying to the end of the input when you run in tail-biting mode, and vice versa?

    VCP2 does not have built-in support for wrapping around to handle the tail biting, so your application must explicitly provide a suitable number of bits before and after the message part that you want to be reliably decoded. (In the thread that Nitin linked to, Shabrin Firdose mentions doing this in the comment from 2014-8-12.  I think TI has an application note about this, but I can't find a link either in E2E or the main web site.)

  • Hi Michael,

    I have gone through the application note about tail-biting and I am attaching the same for your reference.

    Following two approaches I have tried but couldn't get the correct output.

    1.After computing branch metrics, I am copying branch metrics 3 times back-to-back-to-back and giving to VCP2. Frame length also I am giving 3xFramelength. From the output I am pulling out middle portion.

    2. Approach mentioned in application note. (copying (2xK-5xK) of first soft info to last, and vice versa)

    Please let me know if you need any other info regarding this.

    Thanks & Regards,

    Anji Babu

    wi-wimax-tn001-0 0 1-public.pdf