• Communications for Statistical Applications and Methods
• /
• v.12 no.3
• /
• pp.653-657
• /
• 2005

Eigenstructure of a spatial design matrix of Matheron's variogram estimator in $R^1$ is derived. It is shown that the spatial design matrix in $R^1$ with n/2$\le$h < n has a nice spectral decomposition. The mean, variance, and covariance of this estimator are obtained using the eigenvalues of a spatial design matrix. We also found that the lower bound and the upper bound of the normalized Matheron's variogram estimator.

• Journal of IKEEE
• /
• v.27 no.1
• /
• pp.38-44
• /
• 2023

Performance of a STAP(space-time adaptive processing) algorithm highly depends on how closely the estimated covariance matrix(CM) resembles the actual CM by the interference in CUT(cell under test). A STAP has 2 dimensional data structure determined by the number of array elements and the number of transmitting pulses and both numbers are generally not small. Thus, to meet the degree of freedom(DOF) of the CM, a huge amount of training data is required. This paper presents an algorithm to generate virtual training data from small received data, via converting them into the data in spatial frequency-Doppler plane. We theoretically derive where the clutter exist in the plane and present the procedure to implement the proposed algorithm. Finally, with the simulated scenario of small received data, we show the proposed algorithm can improve STAP performance.

• Journal of Korea Multimedia Society
• /
• v.17 no.12
• /
• pp.1446-1452
• /
• 2014

Existing LDA uses the transform matrix that maximizes distance between classes. So we have to convert from an image to one-dimensional vector as training vector. However, in 2D-LDA, we can directly use two-dimensional image itself as training matrix, so that the classification performance can be enhanced about 20% comparing LDA, since the training matrix preserves the spatial information of two-dimensional image. However 2D-LDA uses same calculation schema for transformation matrix and therefore both LDA and 2D-LDA has the heteroscedastic problem which means that the class classification cannot obtain beneficial information of spatial distances of class clusters since LDA uses only data correlation-based covariance matrix of the training data without any reference to distances between classes. In this paper, we propose a new method to apply training matrix of 2D-LDA by using WPS-LDA idea that calculates the reciprocal of distance between classes and apply this weight to between class scatter matrix. The experimental result shows that the discriminating power of proposed 2D-LDA with weighted between class scatter has been improved up to 2% than original 2D-LDA. This method has good performance, especially when the distance between two classes is very close and the dimension of projection axis is low.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2005.02a
• /
• pp.170-177
• /
• 2005

• The Korean Journal of Applied Statistics
• /
• v.23 no.2
• /
• pp.375-382
• /
• 2010

In this study, we propose a robust estimator of the multivariate location parameter by means of the spatial median based on data trimming which extending trimmed mean in the univariate setup. The trimming quantity of this estimator is determined by the bootstrap method, and its covariance matrix is estimated by using the double bootstrap method. This extends the work of Jhun et al. (1993) to the multivariate case. Monte Carlo study shows that the proposed trimmed spatial median estimator yields better efficiency than a spatial median, while its covariance matrix based on double bootstrap overcomes the under-estimating problem occurred on single bootstrap method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.2
• /
• pp.364-370
• /
• 2011

This paper propose a novel method for tracking moving object based on covariance matrix and Riemannian Manifolds. With image backgrounds continuously changed, we use the covariance matrices to extract features for tracking nonrigid object undergoing transformation and deformation. The covariance matrix can make fusion of different types of features and has its small dimension, therefore we enable to handle the spatial and statistical properties as well as the component correlation. The proposed method can estimate the position of the moving object by employing the covariance matrix of object region as a feature vector and comparing the candidate regions. Rimannian Geometry is efficiently adapted to object deformation and change of shape and improve the accuracy by using geodesic distance to predict the estimated position with the minimum distance. The experimental results have shown that the proposed method correctly tracked the moving object.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.6
• /
• pp.1477-1484
• /
• 2015

This paper proposes a robust method for object tracking based on Kalman filters algorithm and covariance matrix. As a feature of the object to be tracked, covariance matrix ensures the continuity of the moving target tracking in the image frames because the covariance is addressed spatial and statistical properties as well as the correlation properties of the features, despite the changes of the form and shape of the target. However, if object moves faster than operation time, real time tracking is difficult. In order to solve the problem, Kalman filters are used to estimate the area of the moving object and covariance matrices as a feature vector are compared with candidate regions within the estimated Kalman window. The results show that the tracking rate of 96.3% achieved using the proposed method.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.5C
• /
• pp.527-539
• /
• 2009

In this paper, we propose an asynchonous ICI (Inter-Cell Interference) mitigation techniques for 3GPP LTE MIMO-OFDMA down-link receiver. An increasing in symbol timing misalignments may occur relative to sychronous network as the result of BS (Base Station) timing differences. Such symbol synchronization errors that exceed the guard interval or the cyclic prefix duration may result in MAI (Multiple Access Interference) for other carriers. In particular, at the cell boundary, this MAI becomes a critical factor, leading to degraded channel throughput and severe asynchronous ICI. Hence, many researchers have investigated the interference mitigation method in the presence of asynchronous ICI and it appears that the knowledge of the SCM (Spatial Covariance Matrix) of the asynchronous ICI plus background noise is an important issue. Generally, it is assumed that the SCM estimated by using training symbols. However, it is difficult to measure the interference statistics for a long time and training symbol is also not appropriate for MIMO-OFDMA system such as LTE. Therefore, a noise reduction method is required to improve the estimation accuracy. Although the conventional time-domain low-pass type weighting method can be effective for noise reduction, it causes significant estimation error due to the spectral leakage in practical OFDM system. Therefore, we propose a time-domain sinc type weighing method which can not only reduce the noise effectively minimizing estimation error caused by the spectral leakage but also implement frequency-domain moving average filter easily. By using computer simulation, we show that the proposed method can provide up to 3dB SIR gain compared with the conventional method.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.2
• /
• pp.120-130
• /
• 2020

This paper solves the problem in underdetermined convolutive mixture by improving the disadvantages of the multichannel nonnegative matrix factorization technique widely used in blind source separation. In conventional researches based on Spatial Covariance Matrix (SCM), each element composed of values such as power gain of single channel and correlation tends to degrade the quality of the separated sources due to high variance. In this paper, level and frequency normalization is performed to effectively cluster the estimated sources. Therefore, we propose a novel SCM and an effective distance function for cluster pairs. In this paper, the proposed SCM is used for the initialization of the spatial model and used for hierarchical agglomerative clustering in the bottom-up approach. The proposed algorithm was experimented using the 'Signal Separation Evaluation Campaign 2008 development dataset'. As a result, the improvement in most of the performance indicators was confirmed by utilizing the 'Blind Source Separation Eval toolbox', an objective source separation quality verification tool, and especially the performance superiority of the typical SDR of 1 dB to 3.5 dB was verified.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.16 no.1
• /
• pp.14-26
• /
• 2011

Surface currents were observed by high-frequency (HF) radars off the Keum River estuary from December 2008 to February 2009. The dataset of observed surface currents had data gaps due to the interference of electromagnetic waves and the deteriorating weather conditions. To fill the data gaps an optimal interpolation procedure was developed. The characteristics of spatial correlation in the surface currents off the Keum River estuary were investigated and the spatial data gaps were filled using the optimal interpolation. Then, the temporal and spatial distribution of the interpolated surface currents and the patterns of interpolation error were examined. The correlation coefficients between the surface currents in the coastal region were higher than 0.7 because tidal currents dominate the surface circulation. The sample data covariance matrix (C), spatially averaged covariance matrix with localization ($C^G_{sm}$) and covariance matrix fitted by an exponential function ($C_{ft}$) were used to interpolate the original dataset. The optimal interpolation filled the data gaps and suppressed the spurious data with spikes in the time series of surface current speed so that the variance of the interpolated time series was smaller than that of the original data. When the spatial data coverage was larger (smaller) than 70% of the region, the interpolation error produced by $C^G_{sm}$ ($C_{ft}$) was smaller compared with that by C.