Showing papers in "Signal Processing in 1984"

TL;DR: A parallel implementation, based on pseudo-QMF filters, is presented as an efficient way to split a signal into equal sub-bands.

TL;DR: A simple algorithm for the evaluation of discrete Fourier transforms (DFT) and discrete cosine transforms (DCT) is presented, which achieves a substantial decrease in the number of additions when compared to currently used FFT algorithms.

TL;DR: A comprehensive analysis of the mean-square learning characteristics of stochastic-descent algorithms is presented, based on the commonly exploited simplifying assumption of stationary independent training vectors.

TL;DR: This dissertation describes the development of a new radiographic reconstruction method designated Ectomography that allows reconstruction of an arbitrarily thick layer of an object using limited viewing angle and estimation and filtering of local image information.

TL;DR: The procedure has been tested by simulation on a computer and has proven to be reliable and robust against the expected perturbing factors such as noise, mismatch of centre frequency or bandwidth, and cross-talk of signals in adjacent frequency channels.

TL;DR: The restoration error and both the deconvoluted signal and its spectrum show clearly the performances of the algorithms and their ability to achieve a spectral extrapolation.

TL;DR: Suboptimal fast transforms are useful substitutes to the optimal Karhunen-Loeve transform (KLT), and the selection of an efficient approximation for the KLT must be done with respect to some performance criterion that might differ from one application to another as mentioned in this paper.

TL;DR: In this article, it was pointed out that the Pseudo-Wigner-Ville distribution may be equivalently regarded as a Short-Time Fourier Transform (STFT).

TL;DR: Fast algorithms are developed which taking into advantage the joint Toeplitz structure, reduce by a factor of two the complexity of existing algorithms for general block ToEplitz matrices.

TL;DR: In this paper, a new structure for the block realization of IIR, 2-D digital filters is proposed, which is based on a matrix representation of 2D convolutions and results in a 2D state variable description with block feedback.

TL;DR: A comparison study of several recently published segmentation procedures, regarding mainly their performance as measured by the ability to detect and localize segment boundaries correctly.

TL;DR: The theoretical groundwork developed for the proof of the eigenvector property of Toeplitz matrices is used to compare Pisarenko's null eigen vector technique with Burg's maximum entropy analysis, and arguments are presented in favor of the former technique.

TL;DR: In this paper, the best plane fit gradient edge detection technique is reexamined and it is shown that this technique for some different mask sizes and different point operator is equivalent to Robert's, Prewitt's, Sobel's and Huckel's gradient techniques.

TL;DR: This method is used for the realization of general, linear m-d filters, with great modularity, and high parallelism, by using known nonsingular matrices and an algorithm for the determination of these coefficients is given and a class of matrices is proposed which leads to simple realizations.

TL;DR: Both approaches indicate that noise and hum make the main contribution to estimation error, and that matched filter estimation is likely to give best estimation accuracy for both P waves and QRS complexes.

TL;DR: In this paper, the equivalence among MEM, AR modeling, and predictive estimation is established, and a new classification method for the least-squares AR spectral estimation methods is introduced: the global and local minimization methods.

TL;DR: It is shown that the increase in sampling time deviations from their synchronous positions increases the efficiency and the proposed method is shown to give a more accurate signal restoration than the traditional one.

TL;DR: La robustesse de la methode est verifiee par son application a une large classe d'images structurees provenant d'experiences astrophysiques et biomedicales.

TL;DR: A new approach to the design of IWSR systems involves a dynamic matching strategy based on the nature of the input speech segment, called signal-dependent matching, which is significantly better than the standard dynamic time warping matching algorithm for confusable as well as nonconfusable vocabulary.

TL;DR: In this paper, a general method for determining the coefficients of maximally flat non-recursive digital filters is presented, and recursive relations with integer coefficients are derived for these coefficients by explicit calculations imitating Crout's method for solving the linear equations obtained by setting the derivatives of the symmetric, linear phase transfer function to zero at w = 0 and w = π.

TL;DR: In this article, the optimal design of Nyquist type signals of M < ∞ baud time duration, which maximizes spectral energy inside a given bandqidth ω and generates no inter-symbol interference is presented.

TL;DR: A simple low-distortion non-linear filter for the restoration of signals degraded by random impulsive noise performs considerably better than a median filter, lowering the reproducing error both in mean-square and maximum-error sense.

TL;DR: Additive noise models for the minimum mean squared error quantization of a random variable are presented and an exact expression for the correlation between the input variable and the quantization noise is derived.

TL;DR: In this work three different pattern compression techniques are compared on the basis of efficiency as well as recognition performance when applied to pattern matching by means of dynamic programming in a speaker dependent context.

TL;DR: Discussion de differentes technique de codage and par modulation d'impulsions codees differentielles, utilisation des effets de masquage du signal de chrominance par le signal de luminance.

TL;DR: Les matrices de Toeplitz apparaissent frequemment dans diverses applications a une ou a plusieurs voies de traitement de signaux telles que l'identification des systemes, le filtrage and la prediction as discussed by the authors.

TL;DR: The kernel for interpolation has been evaluated with the help of polynomial basis functions; this enables efficient implementation of the algorithm.

TL;DR: The author defines the Random Quantization principle and establishes the main properties concerning the moments of the output of a fairly general quantizer with Random Transition Points, and the given expressions of an equivalent transformation highlight the role of the distribution functions of the transition points as a tool of transfer curve synthesis.

TL;DR: A recognition scheme which adapts itself to mild degradations in speech and improves the reliability of recognition significantly is proposed, and techniques which adaptively discriminate between noisy and noise-free parameters by using a selective weighting procedure in the final distance calculation are suggested.