• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.11
• /
• pp.72-82
• /
• 1996

In this paper, newly developed algorithm for enhancing images corrupted by white gaussian noise is proposed. In the method proposed here, image is subdivided into a number of subblocks, and each block is separated into cimponents corresponding to signal and noise subspaces, respectively through the signal subspace method. A clean signal is then estimated form the signal subspace by the adaptive wiener filtering. The decomposition of noisy signal into noise and signal subspaces in is implemented by eigendecomposition of covariance matrix for noisy image, and by performing blockwise KLT (karhunen loeve transformation) using eigenvector. To reduce the perceptual noise level and distortion, wiener filtering is implementd by adaptively adjusting noise level according to activity characteristics of given block. Simulation results show the effectiveness of proposed method. In particular, edge bluring effects are reduced compared to the previous methods.

• Communications for Statistical Applications and Methods
• /
• v.27 no.1
• /
• pp.149-161
• /
• 2020

In this paper we extend the functional principal component analysis for real-valued random functions to the case of Hilbert-space-valued functional random objects. For this, we introduce an autocovariance operator acting on the space of real-valued functions. We establish an eigendecomposition of the autocovariance operator and a Karuhnen-Loève expansion. We propose the estimators of the eigenfunctions and the functional principal component scores, and investigate the rates of convergence of the estimators to their targets. We detail the implementation of the methodology for the cases of compositional vectors and density functions, and illustrate the method by analyzing time-varying population composition data. We also discuss an extension of the methodology to multivariate cases and develop the corresponding theory.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.23 no.6
• /
• pp.1583-1591
• /
• 1998

In this paper, we consider the quasi maximum likelihood(quasi-ML) detector which uses antenna arrays in asynchronous time-varing channels. It is shown that the proposed quasi-ml detector can be regarded as a beamformer followed by a decorrelator: a method based on the eigendecomposition of the correlation matrix of the inverse-filtered signal is proposed to estimate the channel vectors. We also show that the proposed algorithm estimates the channel vector within small mismatch loss in severe propagation environment through computer simulations.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.4A
• /
• pp.399-406
• /
• 2011

Adaptive beamformers, which utilize a priori information on the arrival angle of the desired signal. suppress interferences while maximizing their gains in the desired signal direction. However, if there exist errors in the direction information, they can suffer from severe performance degradation since the desired signal is treated as an interference. A robust adaptive beamforming method is presented which exploits the signal-blocking structure of the Duvall beamformer. The proposed method finds an interference signal space directly from correlations of received signals and then obtains a weight vector such that it is orthogonal to the space. Applying the weight vector to two sub arrays which consist of one less sensors than the original uniform lineal array (ULA), the beamformer efficiently estimates the arrival angle of the desired signal. Its computational complexity is lower than existing methods, which require matrix inversion or eigendecomposition.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.9
• /
• pp.1299-1305
• /
• 2013

In this paper, we investigate a method of reducing the computational cost in speech recognition by dimensionality reduction of MFCC feature vectors. Eigendecomposition of the feature vectors renders linear transformation of the vectors in such a way that puts the vector components in order of variances. The first component has the largest variance and hence serves as the most important one in relevant pattern classification. Therefore, we might consider a method of reducing the computational cost and achieving no degradation of the recognition performance at the same time by dimensionality reduction through exclusion of the least-variance components. Experimental results show that the MFCC components might be reduced by about half without significant adverse effect on the recognition error rate.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.1
• /
• pp.14-20
• /
• 2006

In this paper. a novel technique to set up the initial weights in BSS of delayed mixtures is proposed. After analyzing Eigendecomposition for the correlation matrix of mixing data. the initial weights are set from the Eigenvectors ith delay information. The Proposed setting of initial weighting method for conventional FDICA technique improved the separation Performance. The computer simulation shows that the Proposed method achieves the improved SIR and faster convergence speed of learning curve.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.9
• /
• pp.148-156
• /
• 2014

In the MVDR (minimum variance distortionless response) adaptive array, its performance could be greatly deteriorated in the presence of steering vector errors as the desired signal is treated as an interference. This paper suggests an computationally simple adaptive beamforming method which is robust against these errors. In the proposed method, a minimization problem that is formulated according to the DCB (doubly constrained beamforming) principle is solved to find a solution vector, which is in turn projected onto a subspace to obtain a new steering vector. The minimization problem and the subspace projection are dealt with using some principal eigenpairs, which are obtained using a modified PASTd(projection approximation subspace tracking with deflation). We improve the existing MPASTd(modified PASTd) algorithm such that the computational complexity is reduced. The proposed beamforming method can significantly reduce the complexity as compared with the conventional ones directly eigendecomposing an estimate of the corelation matrix to find all eigenvalues and eigenvectors. Moreover, the proposed method is shown, through simulation, to provide performance improvement over the conventional ones.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.4
• /
• pp.153-159
• /
• 2011

The MUSIC (MUltiple SIgnal Classification) method estimates the directions of arrival (DOAs) of the signals impinging on a sensor array based on the fact that the noise subspace is orthogonal to the signal subspace. In the conventional MUSIC, an estimate of the basis for the noise subspace is obtained by eigendecomposing the sample matrix, which is computationally expensive. In this paper, we present a simple DOA estimation method which finds an estimate of the signal subspace basis directly from the columns of the sample matrix from which the noise power components are removed. DOA estimates are obtained by searching for minimum points of a cost function which is defined using the estimated signal subspace basis. The minimum points are efficiently found through the Brent method which employs parabolic interpolation. Simulation shows that the simple estimation method virtually has the same performance as the complex conventional method based on the eigendecomposition.