I am doing research on a related subject that requires the use of TDOA estimation of two signals in an environment with a great deal of noise and reverberation. If anyone had any expertise on (or links to research regarding) robust passive localization/self-localization, it would be greatly appreciated - thanks!!
$\begingroup$
$\endgroup$
$\begingroup$
$\endgroup$
It is well known that the GCC-PHAT algorithm (based on signals correlation) is robust in the case of reverberation. You can also use preprocessing algorithms to reduce noise.
-
$\begingroup$ Interesting, thanks! - I just found this research which says that Adaptive Eigenvalue Decomposition is even more robust than GCC-PHAT (for both noise and reverb) - does anyone have any insight into how AED works or the difference between GCC-PHAT and AED? link.springer.com/article/10.1155/ASP/2006/26503 $\endgroup$ – Steve Smith Jul 25 '18 at 15:34
