• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 99 subscribers
  • Views 329 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.

Original question:

TDA3XEVM: How is CFAR is performed on Range-Doppler MAP

TDA3XEVM: Does CFAR use the power of RDM cells or the log value of the power?

Yao Yu
Prodigy 80 points
Part Number: TDA3XEVM

In the TI document for TDA3, it was said that CFAR CA detector equation is : CUT > Threshold*NoiseFloor without any other explanations. I have the following questions given each cell of RDM has complex number.

1. Does CUT in the aforementioned equation use the power of the cell under test or  the log value of the power? 

2. If the log is used for CUT, is the noise floor  calculated based on the power of  neighboring cells or the log value of the power?

  • 0
    TI__Mastermind 30245 points

    Thanks for the question! I am following up with our expert on this and would get back to you soon.

    Regards,

    Piyali

  • 0
    TI__Guru 56820 points

    Hi,

    1. Does CUT in the aforementioned equation use the power of the cell under test or  the log value of the power? 

    >>> Energy is first converted into binary log domain

    2. If the log is used for CUT, is the noise floor  calculated based on the power of  neighboring cells or the log value of the powe

    >>> Noise floor calculation happens in log domain.

    Regards,

    Anshu

  • 0 in reply to Anshu Jain
    Prodigy 80 points

     Energy of each bin is first converted into binary log domain and then noise floor calculation happens in log domain. Does it mean that the noise floor is calculated as followed ?

    [(log(x_1)+log(x_2)+...+log(x_N)]/N 

    where x_i denote the power of the i-th reference/training cells.

  • 0 in reply to Yao Yu
    TI__Guru 56820 points

    Hi,

    Yeas you are correct, only difference I see is instead of dividing by N we actually divide by a closest number which is power of 2 and greater than N. This is done because we use shift to handle division.

    Regards,

    Anshu