From the archives

On Approximate Maximum Likelihood Methods for Blind Identification: How to Cope with the Curse of Dimensionality

Historical material. Affiliations, schedules and roles on this page describe their original context.

ConferenceName:  International Workshop on Signal Processing Advances for Wireless Communications

url:  9th IEEE SPAWC (archive lookup)

Edition Number:  9

Date:  June, 2008

Place:  Recife, Brasil

PageStart:  639

PageEnd:  643

Authors:  Steffen BarembruchAurélien GarivierEric Moulines

Abstract: 

We discuss approximate maximum likelihood methods for blind identification and deconvolution. These algorithms are based on particle approximation versions of the EM algorithm. We consider two different methods which differ in the way the posterior distribution of the symbols is computed. The first algorithm is based on a novel particle approximation method of the fixed-interval smoothing whereas the second uses fixed lag smoothing. We compare the two algorithms in a Monte-Carlo experiment; these two methods perform significantly better than the EMVA algorithm, which is considered as the state of the art in this area.

Direct link:  available on IEEE Xplore

Keywords: