• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 101 subscribers
  • Views 594 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.

Two Turbo decoding methods(Max* and Max-Log-MAP) used in TCP2 on DSP TMS C6455

Ben Liu98
Prodigy 20 points

We are using the TCP2 core in TMS C6455 to decode. And there are two decoding methods:Max* and Max-log-MAP. We have checked the reference paper and find the two methods are the same with those in MATLAB. In theory simulation, Max* is better than Max. But in our OFDM system implementation test, Max is much better than Max*. We don't understand why this happen.

In TCP2 datasheet, it says the input for Max* needs channel estimation information and for Max doesn't need SNR estimation. In our system, we do both Channel Estimation and SNR estimation and inputs for Max* and Max are the same. Is that will influence the TCP2 decoding performance? Even that we still can't believe Max-Log_MAP performs better than Max*.

  • 0
    TI__Mastermind 35645 points

    Hi,

    Thanks for your post.

    Ultimately, the decoding of Turbo codes is based on an iterative structure constructed from two MAP decoders and for more details on the decoding algorithm max*-log-MAP and max-log-MAP and its performance, you could refer the SPRABJ0 application report below which would give you an idea how to tune the TCP2 and VCP2 BER performance:

    http://www.ti.com/lit/an/sprabj0/sprabj0.pdf (Refer section 3.1 for the TCP2 decoding algorithm)

    Also. to know more info. on the decoding modes, standalone (SA) or shared processing (SP), you could refer the C6457 TCP2 user guide as  below:

    http://www.ti.com/lit/ug/sprugk1/sprugk1.pdf (refer sections 3 & 4)

    There are examples for the TCP2 in the C6455 CSL. Have you tried running any of those? If not, please try it.

    Thanks & regards,

    Sivaraj K

    -------------------------------------------------------------------------------------------------------

    Please click the Verify Answer button on this post if it answers your question.

    --------------------------------------------------------------------------------------------------------

  • 0 in reply to Sivaraj Kuppuraj
    Prodigy 20 points
    Thanks for your help, but I still can't get understood. Both the report and user guide have been read. In our system, SNR estimation is quite accurate, so we think Max* should outperform Max-Log-MAP. But the results are reverse.
    BTW，I think the figure 7 is wrong in the report. Because it is the same with Figure 8 and it does't agree with its description.