Nova Publishers
My Account Nova Publishers Shopping Cart
HomeBooksSeriesJournalsReference CollectionseBooksInformationSalesImprintsFor Authors
            
  Top » Catalog » Books » Mathematics and Statistics » Kalman Filtering Chapters » My Account  |  Cart Contents  |  Checkout   
Quick Find
  
Use keywords to find the product you are looking for.
Advanced Search
What's New? more
Fetal Alcohol Syndrome: Recognition, Differential Diagnosis and Long-Term Effects
$82.00
Shopping Cart more
0 items
Information
Shipping & Returns
Privacy Notice
Conditions of Use
Contact Us
Notifications more
NotificationsNotify me of updates to Kalman Filtering Approach to Blind Separation of Independent Source Components pp. 225-254
Tell A Friend
 
Tell someone you know about this product.
Kalman Filtering Approach to Blind Separation of Independent Source Components pp. 225-254 $100.00
Authors:  Andreas Galka (Department of Neuropediatrics, University of Kiel, Kiel, Germany) and Tohru Ozaki (Tohoku University in Sendai, Japan)
Abstract:
The problem of extracting a set of independent components from given multivariate
time series data under the assumption of linear instantaneous mixing can be addressed
within a Kalman filtering framework. For this purpose, we introduce a new class of
state space models, the Independent Components State Space Model (IC-SSM). The
resulting algorithm has several attractive features: It takes temporal correlations within
the data into account; it allows for the presence of observation noise; it can deal with
both gaussian and non-gaussian source distributions; it provides a representation for
the main frequencies present in the data; and it succeeds in distinguishing between
dependencies which are introduced by the mixing step, and dependencies which represent
coincidental dependencies resulting from finite time series length. Analysis by
fitting IC-SSM is compared to five well-known algorithms for Independent Component
Analysis (ICA). Through simulations we show that the ICA algorithms, in most
cases, produce estimates of the source components for which the residual mutual information
is too small, as compared to the correct value; this problem does not arise
for IC-SSM. 


Available Options:
Version:
Special Focus Titles
01.Violent Communication and Bullying in Early Childhood Education
02.Cultural Considerations in Intervention with Women and Children Exposed to Intimate Partner Violence
03.Chronic Disease and Disability: The Pediatric Lung
04.Fruit and Vegetable Consumption and Health: New Research
05.Fire and the Sword: Understanding the Impact and Challenge of Organized Islamism. Volume 2

Nova Science Publishers
© Copyright 2004 - 2020

Kalman Filtering Approach to Blind Separation of Independent Source Components pp. 225-254