• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 102 subscribers
  • Views 187 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.

TMS320C5515: Echo cancellation attenuation of VoLIB ECUs.

sekine.kazuma
Prodigy 30 points
Part Number: TMS320C5515

I have a question about the echo canceller attenuation of the VOLIB ECU.
We are currently building a system using a TMS320C5515.
We are planning to use an ECU as a line echo canceller.
How many dB of attenuation is possible for the line echo canceller of the ECU?
Are there any documents (e.g. from actual measurement sites) to prove the attenuation?
Thank you very much in advance.

  • 0
    TI__Mastermind 25685 points

    Hi ,

    Please expect some delay in response due to Public Holiday in US.

    Best Regards,
    Aakash

  • 0
    TI__Mastermind 28848 points

    Hi,

    Please note support on C5000 is limited: https://e2e.ti.com/support/processors-group/processors/f/processors-forum/818771/faq-support-guidance-for-c5000-digital-signal-processors

    I consulted with internal experts on this, please see detailed answered below.

    Regards,
    Frank

    sekine.kazuma said:
    Are there any documents (e.g. from actual measurement sites) to prove the attenuation?

    The only available documentation for VOLIB and ECU are contained in the installed content in the docs folder.

    sekine.kazuma said:
    How many dB of attenuation is possible for the line echo canceller of the ECU?

    Echo attenuation depends on the linearity of the echo path. Essentially all linear echo can be attenuated to the noise floor. Nonlinear echo can be completely killed during single talk. Nonlinear echo attenuation during double talk depends on the configuration. So the answer to this question is not a single number.

    The dB number of echo attenuation depends on linearity and Echo Return Loss (ERL).

    If you have a negative ERL, even if you attenuation echo a lot, you'll still have a lot of residual echo. However, on the other hand, if you already have a high ERL (e.g. 30dB), then there is no much echo to cancel.

    Nonlinearity plays another role. If there is a lot of nonlinearity, you won't get much attenuation from the adaptive filter and you have to rely on NLP. The NLP can completely annihilate the echo, but that would damage near end speech during double talk.

    The ECU meets/exceeds G.168 for reasonable (most common) echo paths.