Paper Info
Title
An efficient entropy rate estimator for complex-valued signal processing: Application to ICA
Abstract
Estimating likelihood or entropy rate is one of the key issues in many signal processing problems. Mutual information rate, which leads to the minimization of entropy rate, provides a natural cost for achieving blind source separation (BSS). In many complex-valued BSS applications, the latent sources are non-Gaussian, noncircular, and possess sample dependence. Consequently, an effective estimator of entropy rate that jointly considers all three properities of the sources is required. In this paper, we propose such an entropy rate estimator that assumes the sources are generated by invertible filters. With this new entropy rate estimator, we propose a complex entropy rate bound minimization algorithm. Simulation results show that the new method exploits all three properties effectively.
Year
DOI
Venue
2014
10.1109/ICASSP.2014.6854799
ICASSP
Keywords
Field
DocType
entropy rate,complex-valued signal processing,independent component analysis,signal processing,likelihood estimation,efficient entropy rate estimator,complex entropy rate bound minimization algorithm,mutual information rate,estimation theory,blind source separation,invertible filter generation,complex-valued bss application,filtering theory,noncircular source,entropy,minimisation,ica,nongaussian source,zirconium,cost function,minimization,vectors
Entropy rate,Transfer entropy,Mathematical optimization,Maximum entropy spectral estimation,Pattern recognition,Entropy maximization,Computer science,Binary entropy function,Artificial intelligence,Principle of maximum entropy,Entropy (information theory),Estimator
Conference
ISSN
Citations 
PageRank 
1520-6149
1
0.36
References 
Authors
0
5
Name
Order
Citations
PageRank
Gengshen Fu11027.82
Ronald Phlypo219016.73
Matthew Anderson326314.64
Xi-Lin Li454734.85
Tülay Adali51690126.40