scholarly journals Adaptive Complex-Valued Independent Component Analysis Based on Second-Order Statistics

2016 ◽  
Vol 2016 ◽  
pp. 1-7 ◽  
Author(s):  
Yanfei Jia ◽  
Xiaodong Yang
Keyword(s):  
Independent Component Analysis ◽  
Cost Function ◽  
Order Statistics ◽  
Component Analysis ◽  
Second Order ◽  
Independent Component ◽  
Second Order Statistics ◽  
The Cost ◽  
Source Signals ◽  
Complex Valued

This paper proposes a two-stage fast convergence adaptive complex-valued independent component analysis based on second-order statistics of complex-valued source signals. The first stage constructs a cost function by extending the real-valued whiten cost function to a complex-valued domain and optimizes the cost function using a complex-valued gradient. The second stage uses the restriction that the pseudocovariance matrix of the separated signal is a diagonal matrix to construct the cost function and the geodesic method is used to optimize the cost function. Compared with other adaptive complex-valued independent component analysis, the proposed method shows a faster convergence rate and smaller error. Computer simulations were performed on synthesized signals and communications signals. The simulation results demonstrate the validity of the proposed algorithm.

An Improved Independent Component Analysis with Reference

2014 ◽  
Vol 667 ◽  
pp. 64-67
Author(s):  
Yan Fei Jia ◽  
Xiao Dong Yang ◽  
Li Yue Xu ◽  
Li Quan Zhao
Keyword(s):  
Independent Component Analysis ◽  
Cost Function ◽  
Newton Method ◽  
Lagrange Function ◽  
Component Analysis ◽  
Convergence Speed ◽  
Independent Component ◽  
Source Signal ◽  
The Cost ◽  
New Algorithms

Independent component analysis with reference is a general framework to incorporate a priori information of interesting source signal into the cost function as constrained terms to form an augmented Lagrange function, and utilizes Newton method to optimize the cost function. It can extract any interesting source signal without extracting all source signals comparing with the traditional Independent component analysis method. In this paper, to accelerate the convergence speed of the Independent component analysis with reference, two improved algorithms are presented. The new algorithms, firstly whiten the observed signals to avoid matrix inverse operation to reduce algorithm complexity, secondly use improved Newton method with fast convergence speed to optimize cost function,in the end deduce the improved Independent component analysis with reference algorithms. Simulation result demonstrates the new algorithms have faster convergence speed with smaller error compared with the original method.

A FAST FIXED-POINT ALGORITHM FOR INDEPENDENT COMPONENT ANALYSIS OF COMPLEX VALUED SIGNALS

2000 ◽  
Vol 10 (01) ◽  
pp. 1-8 ◽  
Author(s):  
ELLA BINGHAM ◽  
AAPO HYVÄRINEN
Keyword(s):  
Fixed Point ◽  
Independent Component Analysis ◽  
Component Analysis ◽  
Independent Component ◽  
Fixed Point Algorithm ◽  
Local Consistency ◽  
Type Algorithm ◽  
Source Signals ◽  
Complex Valued ◽  
Point Type

Separation of complex valued signals is a frequently arising problem in signal processing. For example, separation of convolutively mixed source signals involves computations on complex valued signals. In this article, it is assumed that the original, complex valued source signals are mutually statistically independent, and the problem is solved by the independent component analysis (ICA) model. ICA is a statistical method for transforming an observed multidimensional random vector into components that are mutually as independent as possible. In this article, a fast fixed-point type algorithm that is capable of separating complex valued, linearly mixed source signals is presented and its computational efficiency is shown by simulations. Also, the local consistency of the estimator given by the algorithm is proved.

Separation and Reproduction of Mixed Images Using 2-D Complex Wavelet Transform

2010 ◽  
Vol 36 ◽  
pp. 466-475
Author(s):  
Tsutomu Matsuura ◽  
Amirul Faiz ◽  
Kouji Kiryu
Keyword(s):  
Image Processing ◽  
Wavelet Transform ◽  
Independent Component Analysis ◽  
Component Analysis ◽  
Independent Component ◽  
Complex Wavelet Transform ◽  
Time Frequency ◽  
Frequency Information ◽  
Complex Wavelet ◽  
Complex Valued

The differences method between 1-D wavelet transform and 2-D wavelet transform in image processing is discussed. Both proposed method uses the quotient of complex valued time-frequency information of observed signals to detect the number of sources. No less number of observed signals than the detected number of sources is needed to separate sources. The assumption on sources is quite general independence in the time-frequency plane, which is different from that of independent component analysis. Using the same given Algorithm and parameters for both method, the result on separated images are compared.

Sign in / Sign up

Export Citation Format

Share Document