We examine sequential algorithms and formulate a sequential-time postulate, an abstractstatepostulate, and a bounded-exploration postulate. Analysis of the postulates leads us tothe notion of sequential abstract-state...
详细信息
Finding a feasible point that satisfies a set of constraints is a common task in scientific computing;examples are the linear feasibility problem and the convex feasibility problem. Finitely convergent sequential algo...
详细信息
Finding a feasible point that satisfies a set of constraints is a common task in scientific computing;examples are the linear feasibility problem and the convex feasibility problem. Finitely convergent sequential algorithms can be used for solving such problems;an example of such an algorithm is ART3, which is defined in such a way that its control is cyclic in the sense that during its execution it repeatedly cycles through the given constraints. Previously we found a variant of ART3 whose control is no longer cyclic, but which is still finitely convergent and in practice usually converges faster than ART3. In this article we propose a general methodology for automatic transformation of finitely convergent sequential algorithms in such a way that (1) finite convergence is retained, and (2) the speed of convergence is improved. The first of these properties is proven by mathematical theorems, the second is illustrated by applying the algorithms to a practical problem.
This paper presents a novel speech feature enhancement technique based on a probabilistic, nonlinear acoustic environment model that effectively incorporates the phase relationship (hence phase sensitive) between the ...
详细信息
This paper presents a novel speech feature enhancement technique based on a probabilistic, nonlinear acoustic environment model that effectively incorporates the phase relationship (hence phase sensitive) between the clean speech and. the corrupting noise in the acoustic distortion process. The core of the enhancement algorithm is the MMSE (minimum mean square error) estimator for the log Mel power spectra of clean speech based on the phase-sensitive environment model, using highly efficient single-point, second-order Taylor series expansion to approximate the joint probability of clean and noisy speech modeled as a multivariate Gaussian. Since a noise estimate is required by the MMSE estimator, a high-quality, sequential noise estimation algorithm is also developed and presented. Both the noise estimation and speech feature enhancement algorithms are evaluated on the Aurora2 task of connected digit recognition. Noise-robust speech recognition results demonstrate that the new acoustic environment model which takes into account, the relative phase in speech and noise mixing is superior to the earlier environment model which discards the phase under otherwise identical experimental conditions. The results also show that the sequential MAP (maximum a posteriori) learning for noise estimation is better. than the sequential ML (maximum likelihood) learning, both evaluated under the identical phase-sensitive MMSE enhancement condition.
A receiver consisting of a whitened matched filter (WMF) followed by a sequential algorithm (SA) had been developed previously for intersymbol interference (ISI) channels. The channel-WMF system can be modeled as a fi...
详细信息
A receiver consisting of a whitened matched filter (WMF) followed by a sequential algorithm (SA) had been developed previously for intersymbol interference (ISI) channels. The channel-WMF system can be modeled as a finite-state machine (FSM). This paper presents an analysis of SA computational complexity. To determine the computational complexity, the FSM can be interpreted as a special convolutional encoder followed by a binary symbol to Q-ary symbol mapping. It follows that the computational distribution is Pareto, and that there exists a computational cutoff rate R(comp). For the uncoded data considered, the rate is fixed and the R(comp) criterion translates into a signal-to-noise ratio (SNR) criterion, i.e., for bounded computation per node, the SNR should exceed SNR(comp). An upper bound SNR(comp)' on SNR(comp) is found analytically by assuming a uniform input distribution. Arimoto's iteration equations are adapted to find the true SNR(comp) numerically.
A vector sequential sequence estimator is proposed for multiple-channel transmission systems with both intersymbol interference (ISI) and interchannel interference (ICI). Both finite ISI-ICI and infinite ISI-ICI are c...
详细信息
A vector sequential sequence estimator is proposed for multiple-channel transmission systems with both intersymbol interference (ISI) and interchannel interference (ICI). Both finite ISI-ICI and infinite ISI-ICI are considered. The estimator consists of a multiple-dimensional whitened matched filter and a vector sequential decoder. The metric of the sequential algorithm is derived and the algorithm's performance analysis is given. Computer simulation results for a two-dimensional finite ISI-ICI channel and a two-dimensional infinite ISI-ICI channel are presented. Analysis and simulation show that the symbol error probability of the vector sequential algorithm is essentially the same as maximum likelihood sequence estimation via the vector Viterbi algorithm (VA), while its average computational complexity is much less, although computation per symbol is a random variable with the Pareto distribution. There exists a signal-to-noise ratio SNR(comp) above which the ensemble average computation is bounded. An upper bound on the SNR(comp) is found.
Determining an optimal design for estimation of parameters of a class of complex models expected to be built at a minimum cost is a growing trend in science and engineering. We adopt a scale-bias adjustment migration ...
详细信息
Determining an optimal design for estimation of parameters of a class of complex models expected to be built at a minimum cost is a growing trend in science and engineering. We adopt a scale-bias adjustment migration strategy for integrating base and new models based on similar nature underlying processes. Further, we propose a Bayesian sequential algorithm for obtaining the statistically most informative data about the migrated model for use in parameter estimation. The benefits of the proposed strategy over traditional approaches presented in recent reported work are demonstrated using Monte Carlo simulations. (C) 2014 Elsevier Ltd. All rights reserved.
Multiscale noise tuning stochastic resonance (MSTSR) has been proved to be an effective method for enhanced fault diagnosis by taking advantage of noise to detect the incipient faults of the bearings and gearbox. This...
详细信息
Multiscale noise tuning stochastic resonance (MSTSR) has been proved to be an effective method for enhanced fault diagnosis by taking advantage of noise to detect the incipient faults of the bearings and gearbox. This paper addresses a sequential algorithm for the MSTSR method to detect the train bearing faults in an embedded system through the acoustic signal analysis. Specifically, the energy operator, digital filter array, and fourth rank Runge-Kutta equation methods are designed to realize the signal demodulation, multiscale noise tuning, and bistable stochastic resonance in sequence. The merit of the sequential algorithm is that it reduces the memory consumption and decreases the computation complexity, so that it can be efficiently implemented in the embedded system based on a low-cost, low-power hardware platform. After the sequential algorithm, the real-valued fast Fourier transform is used to calculate the power spectrum of the analyzed signal. The proposed method has been verified in algorithm performance and hardware implementation by three kinds of practical acoustic signals from defective train bearings. An enhanced performance of the proposed fault diagnosis method is confirmed as compared with several traditional methods, and the hardware performance is also validated.
Principal component analysis (PCA) and Minor component analysis (MCA) are similar but have different dynamical performances. Unexpectedly, a sequential extraction algorithm for MCA proposed by Luo and Unbehauen [11] d...
详细信息
Principal component analysis (PCA) and Minor component analysis (MCA) are similar but have different dynamical performances. Unexpectedly, a sequential extraction algorithm for MCA proposed by Luo and Unbehauen [11] does not work for MCA, while it works for PCA. We propose a different sequential-addition algorithm which works for MCA. We also show a conversion mechanism by which any PCA algorithms are converted to dynamically equivalent MCA algorithms and vice versa.
A method for compressing binary images is monochromatic pattern substitution. Monochromatic rectangles inside the image are detected and compressed by a variable length code. Such method has no relevant loss of compre...
详细信息
ISBN:
(纸本)9788001048702
A method for compressing binary images is monochromatic pattern substitution. Monochromatic rectangles inside the image are detected and compressed by a variable length code. Such method has no relevant loss of compression effectiveness if the image is partitioned into up to a thousand blocks and each block is compressed independently. Therefore, it can be implemented in parallel on both small and large scale arrays of processors with distributed memory and no interconnections. We show in this paper that such method has a speed-up if applied sequentially to the partitioned image. Experimental results show that the speed-up happens if the image is partitioned into up to 256 blocks and sequentially each block is compressed independently. It follows that the sequential speed-up can also be applied to a parallel implementation on a small scale system.
We describe the construction of accurate panoramic mosaics from multiple images taken with a rotating camera, or alternatively of a planar scene. The novelty of the approach lies in (i) the transfer of photogrammetric...
详细信息
We describe the construction of accurate panoramic mosaics from multiple images taken with a rotating camera, or alternatively of a planar scene. The novelty of the approach lies in (i) the transfer of photogrammetric bundle adjustment techniques to mosaicing;(ii) a new representation of image line measurements enabling the use of lines in camera self-calibration, including computation of the radial and other non-linear distortion;and (iii) the application of the variable state dimension filter to obtain efficient sequential updates of the mosaic as each image is added. We demonstrate that our method achieves better results than the alternative approach of optimising over pairs of images. (C) 2002 Elsevier Science B.V. All rights reserved.
暂无评论