IJCNN'01. International Joint Conference on Neural Networks. Proceedings (Cat. No.01CH37222)
DOI: 10.1109/ijcnn.2001.939479
|View full text |Cite
|
Sign up to set email alerts
|

Blind source recovery: algorithms for static and dynamic environments

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
5

Citation Types

2
47
0

Publication Types

Select...
3
2
2

Relationship

2
5

Authors

Journals

citations
Cited by 18 publications
(49 citation statements)
references
References 8 publications
2
47
0
Order By: Relevance
“…Unlike the transfer function models, the state-space provides an efficient internal description of a system. Moreover, there are various possible equivalent state space realizations for a system, and thus recovery of original sources can be achieved independent from (and even in the absence of) environment identifiability [5]. Further, The inverse for a state space representation is easily derived subject to the "invertibility" of the instantaneous relational mixing matrix between input-output.…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…Unlike the transfer function models, the state-space provides an efficient internal description of a system. Moreover, there are various possible equivalent state space realizations for a system, and thus recovery of original sources can be achieved independent from (and even in the absence of) environment identifiability [5]. Further, The inverse for a state space representation is easily derived subject to the "invertibility" of the instantaneous relational mixing matrix between input-output.…”
Section: Introductionmentioning
confidence: 99%
“…Furthermore, the asymptotic convergence of the algorithm is slower as compared to the minimum phase systems, this can be attributed to the transients induced in the algorithm due to bi-directional nature of the algorithm [8] The state space notion provides a compact representation, which is capable of handling both time delayed and filtered versions of signals in an organized manner [5,6,7]. Unlike the transfer function models, the state-space provides an efficient internal description of a system.…”
Section: Introductionmentioning
confidence: 99%
“…The state space notion provides a compact representation, capable of handling both time delayed and filtered versions of signals in an organized manner [2,3,6]. Unlike the transfer function models of standard dynamic filters, the use of the state-space can result in several generalized, equivalent and efficient internal descriptions of a system.…”
Section: Introductionmentioning
confidence: 99%
“…The performance of BSR algorithms strongly depends on the choice of an appropriate score function which appears as an element wise acting non-linearity on the output signals [2][3][4][5][6]. For a particular problem, the optimal score function depends on the distribution of the original source signals which are unknown in a blind scenario.…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation