Home About us Contact
|
Home About us Contact | ||
|
|
||
Convergence Speed (convergence + speed)
Selected AbstractsSemi-blind fast equalization of QAM channels using concurrent gradient-Newton CMA and soft decision-directed schemeINTERNATIONAL JOURNAL OF ADAPTIVE CONTROL AND SIGNAL PROCESSING, Issue 6 2010S. Chen Abstract This contribution considers semi-blind adaptive equalization for communication systems that employ high-throughput quadrature amplitude modulation signalling. A minimum number of training symbols, approximately equal to the dimension of the equalizer, are first utilized to provide a rough initial least-squares estimate of the equalizer's weight vector. A novel gradient-Newton concurrent constant modulus algorithm and soft decision-directed scheme are then applied to adapt the equalizer. The proposed semi-blind adaptive algorithm is capable of converging fast and accurately to the optimal minimum mean-square error equalization solution. Simulation results obtained demonstrate that the convergence speed of this semi-blind adaptive algorithm is close to that of the training-based recursive least-square algorithm. Copyright © 2009 John Wiley & Sons, Ltd. [source] Fractionally spaced blind equalization with low-complexity concurrent constant modulus algorithm and soft decision-directed schemeINTERNATIONAL JOURNAL OF ADAPTIVE CONTROL AND SIGNAL PROCESSING, Issue 6 2005S. Chen Abstract The paper proposes a low-complexity concurrent constant modulus algorithm (CMA) and soft decision-directed (SDD) scheme for fractionally spaced blind equalization of high-order quadrature amplitude modulation channels. We compare our proposed blind equalizer with the recently introduced state-of-art concurrent CMA and decision-directed (DD) scheme. The proposed CMA+SDD blind equalizer is shown to have simpler computational complexity per weight update, faster convergence speed, and slightly improved steady-state equalization performance, compared with the existing CMA+DD blind equalizer. Copyright © 2004 John Wiley & Sons, Ltd. [source] Practical implementation of multichannel adaptive filters based on FTF and AP algorithms for active controlINTERNATIONAL JOURNAL OF ADAPTIVE CONTROL AND SIGNAL PROCESSING, Issue 2-3 2005Alberto González Abstract In this paper, multichannel affine projection (AP) algorithms and fast transversal filters (FTF) are introduced for active noise control. A comparative practical study of the mentioned algorithms with the filtered-X LMS (F-XLMS) and the recursive least squares (RLS) is presented for multichannel systems. This study is based on simulations using real data and is mainly focused on: their computational cost and memory load, their convergence properties, their stability and their ability to create quiet zones around listener ears. Simulations show that algorithms based on FTF exhibit a good trade-off between computational cost and convergence speed. On the other hand, those based on RLS are slightly faster but they present higher computational load and stability problems in their practical implementation. It has also been observed that algorithms based on low order AP algorithms present less computational cost than the FTF-based ones but a slightly slower convergence speed. Therefore these algorithms show a desirable behaviour and versatility for practical applications. Finally, results obtained in a real-time multichannel system validate the use of AP algorithms in practical applications as an alternative to the classical multichannel F-XLMS since they provide meaningful attenuation levels, lower convergence time and similar computational cost. Additionally, as simulations indicated, AP algorithm performance can be easily improved increasing its projection order and using fast versions. Copyright © 2004 John Wiley & Sons, Ltd. [source] The Gauss-Seidel fast affine projection algorithm for multichannel active noise control and sound reproduction systemsINTERNATIONAL JOURNAL OF ADAPTIVE CONTROL AND SIGNAL PROCESSING, Issue 2-3 2005Martin Bouchard Abstract In the field of adaptive filtering, the fast implementations of affine projection algorithms are known to provide a good tradeoff between convergence speed and computational complexity. Such algorithms have recently been published for multichannel active noise control systems. Previous work reported that these algorithms can outperform more complex recursive least-squares algorithms when noisy plant models are used in active noise control systems. This paper proposes a new fast affine projection algorithm for multichannel active noise control or sound reproduction systems, based on the Gauss,Seidel solving scheme. The proposed algorithm has a lower complexity than the previously published algorithms, with the same convergence speed and the same good performance with noisy plant models, and a potential for better numerical stability. It provides the best performance/cost ratio. Details of the algorithm and its complexity are presented in the paper, with simulation results to validate its performance. Copyright © 2004 John Wiley & Sons, Ltd. [source] Blind separation of convolutive mixtures of cyclostationary signalsINTERNATIONAL JOURNAL OF ADAPTIVE CONTROL AND SIGNAL PROCESSING, Issue 3 2004Wenwu Wang Abstract An adaptive blind source separation algorithm for the separation of convolutive mixtures of cyclostationary signals is proposed. The algorithm is derived by applying natural gradient iterative learning to a novel cost function which is defined according to the wide sense cyclostationarity of signals and can be deemed as a new member of the family of natural gradient algorithms for convolutive mixtures. A method based on estimating the cycle frequencies required for practical implementation of the proposed algorithm is presented. The efficiency of the algorithm is supported by simulations, which show that the proposed algorithm has improved performance for the separation of convolved cyclostationary signals in terms of convergence speed and waveform similarity measurement, as compared to the conventional natural gradient algorithm for convolutive mixtures. Copyright © 2004 John Wiley & Sons, Ltd. [source] Natural gradient algorithm for neural networks applied to non-linear high power amplifiers,INTERNATIONAL JOURNAL OF ADAPTIVE CONTROL AND SIGNAL PROCESSING, Issue 8 2002H. Abdulkader Abstract This paper investigates the processing techniques for non-linear high power amplifiers (HPA) using neural networks (NNs). Several applications are presented: Identification and Predistortion of the HPA. Various Neural Network structures are proposed to identify and predistort the HPA. Since a few decades, NNs have shown excellent performance in solving complex problems (like classification, recognition, etc.) but usually they suffer from slow convergence speed. Here, we propose to use the natural gradient instead of the classical ordinary gradient in order to enhance the convergence properties. Results are presented concerning identification and predistortion using classical and natural gradient. Practical implementations issues are given at the end of the paper. Copyright © 2002 John Wiley & Sons, Ltd. [source] Properties and performance of orthogonal neural network in function approximationINTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS, Issue 12 2001Chieh F. Sher Backpropagation neural network has been applied successfully to solving uncertain problems in many fields. However, unsolved drawbacks still exist such as the problems of local minimum, slow convergence speed, and the determination of initial weights and the number of processing elements. In this paper, we introduce a single-layer orthogonal neural network (ONN) that is developed based on orthogonal functions. Since the processing elements are orthogonal to one another and there is no local minimum of the error function, the orthogonal neural network is able to avoid the above problems. Among the five existing orthogonal functions, Legendre polynomials and Chebyshev polynomials of the first kind have the properties of recursion and completeness. They are the most suitable to generate the neural network. Some typical examples are given to show their performance in function approximation. The results show that ONN has excellent convergence performance. Moreover, ONN is capable of approximating the mathematic model of backpropagation neural network. Therefore, it should be able to be applied to various applications that backpropagation neural network is suitable to solve. © 2001 John Wiley & Sons, Inc. [source] Comparison of image reconstruction by using near-field and far-field data for an imperfect conductorINTERNATIONAL JOURNAL OF RF AND MICROWAVE COMPUTER-AIDED ENGINEERING, Issue 2 2001Chien-Ching Chiu Abstract Image reconstruction by using near-field and far-field data for an imperfectly conducting cylinder is investigated. A conducting cylinder of unknown shape and conductivity scatters the incident wave in free space and the scattered near and far fields are measured. By using measured fields, the imaging problem is reformulated into an optimization problem and solved by the genetic algorithm. Numerical results show that the convergence speed and final reconstructed results by using near-field data are better than those obtained by using far-field data. This work provides both comparative and quantitative information. © 2001 John Wiley & Sons, Inc. Int J RF and Microwave CAE 11: 69,73, 2001. [source] Nonlinear Cointegration Relationships Between Non-Life Insurance Premiums and Financial MarketsJOURNAL OF RISK AND INSURANCE, Issue 3 2009Fredj Jawadi The aim of this article is to study the adjustment dynamics of the non-life insurance premium (NLIP) and test its dependence to the financial markets in five countries (Canada, France, Japan, the United Kingdom, and the United States). First, we justify the linkage between the insurance and the financial markets by the underwriting cycle theory and financial models of insurance pricing. Second, we examine the relationship between the NLIP, the interest rate, and the stock price using the recent developments of nonlinear econometrics. We use threshold cointegration models: the switching transition error correction models (STECM). We show that STECM perform better than a linear error correction model (LECM) to reproduce the NLIP dynamics. Our empirical results show that the adjustment of the NLIP in France, Japan, and the United States is rather discontinuous, asymmetrical, and nonlinear. Moreover, we suggest a strong evidence of significant linkages between insurance and financial markets, show two regimes for the NLIP, and find that the NLIP adjustment toward equilibrium is time varying with a convergence speed that varies according to the insurance disequilibrium size. [source] The effect of overall discretization scheme on Jacobian structure, convergence rate, and solution accuracy within the local rectangular refinement methodNUMERICAL LINEAR ALGEBRA WITH APPLICATIONS, Issue 8 2001Beth Anne V. Bennett Abstract The local rectangular refinement (LRR) solution-adaptive gridding method automatically produces orthogonal unstructured adaptive grids and incorporates multiple-scale finite differences to discretize systems of elliptic governing partial differential equations (PDEs). The coupled non-linear discretized equations are solved simultaneously via Newton's method with a Bi-CGSTAB linear system solver. The grids' unstructured nature produces a nonstandard sparsity pattern within the Jacobian. The effects of two discretization schemes (LRR multiple-scale stencils and traditional single-scale stencils) on Jacobian bandwidth, convergence speed, and solution accuracy are studied. With various point orderings, for two simple problems with analytical solutions, the LRR multiple-scale stencils are seen to: (1) produce Jacobians of smaller bandwidths than those resulting from the traditional single-scale stencils; (2) lead to significantly faster Newton's method convergence than the single-scale stencils; and (3) produce more accurate solutions than the single-scale stencils. The LRR method, including the LRR multiple-scale stencils, is finally applied to an engineering problem governed by strongly coupled, highly non-linear PDEs: a steady-state lean Bunsen flame with complex chemistry, multicomponent transport, and radiation modeling. Very good agreement is observed between the computed flame height and previously published experimental data. Copyright © 2001 John Wiley & Sons, Ltd. [source] On-line learning for very large data setsAPPLIED STOCHASTIC MODELS IN BUSINESS AND INDUSTRY, Issue 2 2005Léon Bottou Abstract The design of very large learning systems presents many unsolved challenges. Consider, for instance, a system that ,watches' television for a few weeks and learns to enumerate the objects present in these images. Most current learning algorithms do not scale well enough to handle such massive quantities of data. Experience suggests that the stochastic learning algorithms are best suited to such tasks. This is at first surprising because stochastic learning algorithms optimize the training error rather slowly. Our paper reconsiders the convergence speed in terms of how fast a learning algorithm optimizes the testing error. This reformulation shows the superiority of the well designed stochastic learning algorithm. Copyright © 2005 John Wiley & Sons, Ltd. [source] | |||||||||||||