I was going through this paper and the author mentioned that we can prove the following using the Matrix Inversion Lemma (AKA Woodburry's Matrix Inversion Identity):
Using matrix inversion lemma we can show that
$$w(k+1) = w(k) + \mu(k)e_p(k)x(k)$$
and $$e_p(k) = d(k) - x^T(k)w(k+1)$$
where $e_p(k)$ is the a posteriori error, is equivalent to the NLMS coefficient update equations $eq.(1)$ and $eq.(2)$ which employs a priori error in its update equation.

The way I did it, I did not use the Matrix Inversion Lemma and still proved the equivalence I want to know how to do it with the Matrix Inversion Lemma (which apparently is giving me a hard time). I would be glad if anyone could help me out here. Even a hint would do.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.