Abstract: In the present paper, we propose numerical methods for solving the Stein equation AXC - X - D = 0 where the matrix A is large and sparse. Such problems appear in discrete-time control problems, filtering and image restoration. We consider the case where the matrix D is of full rank and the case where D is factored as a product of two matrices. The proposed methods are Krylov subspace methods based on the block Arnoldi algorithm. We give theoretical results and we report some numerical experiments.
Abstract: Restarted GMRES methods augmented with approximate eigenvectors are widely used for solving large sparse linear systems. Recently a new scheme of augmenting with error approximations is proposed. The main aim of this paper is to develop a restarted GMRES method augmented with the combination of harmonic Ritz vectors and error approximations. We demonstrate that the resulted combination method can gain the advantages of two approaches: (i) effectively deflate the small eigenvalues in magnitude that may hamper the convergence of the method and (ii) partially recover the global optimality lost due to restarting. The effectiveness and efficiency of the new method are demonstrated through various numerical examples.
Abstract: In this work the opportunity of construction of the
qualifiers for face-recognition systems based on conjugation criteria
is investigated. The linkage between the bipartite conjugation, the
conjugation with a subspace and the conjugation with the null-space
is shown. The unified solving rule is investigated. It makes the
decision on the rating of face to a class considering the linkage
between conjugation values. The described recognition method can
be successfully applied to the distributed systems of video control
and video observation.
Abstract: In this paper a novel method for multiple one dimensional real valued sinusoidal signal frequency estimation in the presence of additive Gaussian noise is postulated. A computationally simple frequency estimation method with efficient statistical performance is attractive in many array signal processing applications. The prime focus of this paper is to combine the subspace-based technique and a simple peak search approach. This paper presents a variant of the Propagator Method (PM), where a collaborative approach of SUMWE and Propagator method is applied in order to estimate the multiple real valued sine wave frequencies. A new data model is proposed, which gives the dimension of the signal subspace is equal to the number of frequencies present in the observation. But, the signal subspace dimension is twice the number of frequencies in the conventional MUSIC method for estimating frequencies of real-valued sinusoidal signal. The statistical analysis of the proposed method is studied, and the explicit expression of asymptotic (large-sample) mean-squared-error (MSE) or variance of the estimation error is derived. The performance of the method is demonstrated, and the theoretical analysis is substantiated through numerical examples. The proposed method can achieve sustainable high estimation accuracy and frequency resolution at a lower SNR, which is verified by simulation by comparing with conventional MUSIC, ESPRIT and Propagator Method.
Abstract: This paper deals with the localization of the wideband sources. We develop a new approach for estimating the wide band sources parameters. This method is based on the high order statistics of the recorded data in order to eliminate the Gaussian components from the signals received on the various hydrophones.In fact the noise of sea bottom is regarded as being Gaussian. Thanks to the coherent signal subspace algorithm based on the cumulant matrix of the received data instead of the cross-spectral matrix the wideband correlated sources are perfectly located in the very noisy environment. We demonstrate the performance of the proposed algorithm on the real data recorded during an underwater acoustics experiments.
Abstract: This paper presents a text clustering system developed based on a k-means type subspace clustering algorithm to cluster large, high dimensional and sparse text data. In this algorithm, a new step is added in the k-means clustering process to automatically calculate the weights of keywords in each cluster so that the important words of a cluster can be identified by the weight values. For understanding and interpretation of clustering results, a few keywords that can best represent the semantic topic are extracted from each cluster. Two methods are used to extract the representative words. The candidate words are first selected according to their weights calculated by our new algorithm. Then, the candidates are fed to the WordNet to identify the set of noun words and consolidate the synonymy and hyponymy words. Experimental results have shown that the clustering algorithm is superior to the other subspace clustering algorithms, such as PROCLUS and HARP and kmeans type algorithm, e.g., Bisecting-KMeans. Furthermore, the word extraction method is effective in selection of the words to represent the topics of the clusters.
Abstract: A subsea hydrocarbon production system can undergo planned and unplanned shutdowns during the life of the field. The thermal FEA is used to simulate the cool down to verify the insulation design of the subsea equipment, but it is also used to derive an acceptable insulation design for the cold spots. The driving factors of subsea analyses require fast responding and accurate models of the equipment cool down. This paper presents cool down analysis carried out by a Krylov subspace reduction method, and compares this approach to the commonly used FEA solvers. The model considered represents a typical component of a subsea production system, a closed valve on a dead leg. The results from the Krylov reduction method exhibits the least error and requires the shortest computational time to reach the solution. These findings make the Krylov model order reduction method very suitable for the above mentioned subsea applications.
Abstract: In recent years, a number of works proposing the
combination of multiple classifiers to produce a single
classification have been reported in remote sensing literature. The
resulting classifier, referred to as an ensemble classifier, is
generally found to be more accurate than any of the individual
classifiers making up the ensemble. As accuracy is the primary
concern, much of the research in the field of land cover
classification is focused on improving classification accuracy. This
study compares the performance of four ensemble approaches
(boosting, bagging, DECORATE and random subspace) with a
univariate decision tree as base classifier. Two training datasets,
one without ant noise and other with 20 percent noise was used to
judge the performance of different ensemble approaches. Results
with noise free data set suggest an improvement of about 4% in
classification accuracy with all ensemble approaches in
comparison to the results provided by univariate decision tree
classifier. Highest classification accuracy of 87.43% was achieved
by boosted decision tree. A comparison of results with noisy data
set suggests that bagging, DECORATE and random subspace
approaches works well with this data whereas the performance of
boosted decision tree degrades and a classification accuracy of
79.7% is achieved which is even lower than that is achieved (i.e.
80.02%) by using unboosted decision tree classifier.
Abstract: Studies in neuroscience suggest that both global and
local feature information are crucial for perception and recognition of
faces. It is widely believed that local feature is less sensitive to
variations caused by illumination, expression and illumination. In
this paper, we target at designing and learning local features for face
recognition. We designed three types of local features. They are
semi-global feature, local patch feature and tangent shape feature.
The designing of semi-global feature aims at taking advantage of
global-like feature and meanwhile avoiding suppressing AdaBoost
algorithm in boosting weak classifies established from small local
patches. The designing of local patch feature targets at automatically
selecting discriminative features, and is thus different with traditional
ways, in which local patches are usually selected manually to cover
the salient facial components. Also, shape feature is considered in
this paper for frontal view face recognition. These features are
selected and combined under the framework of boosting algorithm
and cascade structure. The experimental results demonstrate that the
proposed approach outperforms the standard eigenface method and
Bayesian method. Moreover, the selected local features and
observations in the experiments are enlightening to researches in
local feature design in face recognition.
Abstract: DS-CDMA system is well known wireless
technology. This system suffers from MAI (Multiple Access
Interference) caused by Direct Sequence users. Multi-User Detection
schemes were introduced to detect the users- data in presence of
MAI. This paper focuses on linear multi-user detection schemes used
for data demodulation. Simulation results depict the performance of
three detectors viz-conventional detector, Decorrelating detector and
Subspace MMSE (Minimum Mean Square Error) detector. It is seen
that the performance of these detectors depends on the number of
paths and the length of Gold code used.