I am trying to implement an algorithm proposed in an old research paper by Ephraim and Malah for my DSP course project. I hope someone can help me with this difficult question. For the clean signal Amplitude Estimator, they used the following equation:

enter image description here

I understand most of the variables in the equation except the Bessel function stuff. Why did they use a Bessel function in here? and most importantly, how can I implement it in Matlab?


1 Answer 1


The Bessel function just turns up as a result of the MMSE estimation problem given the assumed statistical model (cf. Appendix A of the paper). This modified Bessel function is implemented in Matlab as besseli.

  • $\begingroup$ Thank you very much for replying to me Matt. I looked into Appendix A and also read about Bessli function in general and in matlab in specific, but I didn't know how to use it for zero-order and first order. Both have the same index of (vk/2), so in matlab I did besseli(0,vk/2) and besseli(1,vk/2) for I0(vk/2) and I1(vk/2) respectively! Is this correct?!!! Please let me know $\endgroup$
    – Mona
    Commented Nov 25, 2014 at 21:24
  • $\begingroup$ @Mona: Yes, that's correct. $\endgroup$
    – Matt L.
    Commented Nov 25, 2014 at 21:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.