Independent Component Analysis: A Review |
Author(s): |
| Pratika Tiwari , PIET, Limda; Prof. Mitul Patel , PIET, Limda |
Keywords: |
| Independent Component Analysis, Cocktail Party, Blind Source Separation, Signal Separation, Whitening, Centering |
Abstract |
|
ICA is solution of the fundamental problem in neural network research, also in other areas, that is to find a proper representation of multivariant data i.e. random vectors. This is linear transformation of original data. We can say that each component is representation of linear combination of its original variables. There are very well-known transformation methods like principal component analysis, projection pursuit and factor analysis. So Independent Component Analysis (ICA) is a method with help of which we can have a linear representation of nongaussian data so that the components are statistically independent. So, in this paper we see the basic theory and application of ICA. |
Other Details |
|
Paper ID: IJSRDV3I31618 Published in: Volume : 3, Issue : 3 Publication Date: 01/06/2015 Page(s): 3699-3701 |
Article Preview |
|
|
|
|
