# IEEE Transactions on Signal Processing

## Filter Results

Displaying Results 1 - 25 of 36
• ### Integral inequality bounding the weighted absolute deviation of an n-dimensional function

Publication Year: 1992, Page(s):973 - 975
Cited by:  Papers (2)
| | PDF (224 KB)

The author states and proves an integral inequality that bounds the weighted integrated absolute deviation of a differentiable n-dimensional real function over an interval, relative to any value the function takes within the interval. Examples illustrate the utility of the inequality. In particular, the inequality is shown to be applicable to certain set-theoretic signal restoration algor... View full abstract»

• ### Maximum entropy regularization in inverse synthetic aperture radar imagery

Publication Year: 1992, Page(s):969 - 973
Cited by:  Papers (12)
| | PDF (376 KB)

The method of maximum entropy is applied to the regularization of inverse synthetic aperture radar (ISAR) image reconstructions. This is accomplished by considering an ensemble of images with associated allowed' probability density functions. Instead of directly considering the solution' to be an image, the author takes it to be the a posteriori probability density found by minimizing a regulari... View full abstract»

• ### A comment on `Decomposition of the LPC excitation using the zinc basic functions' [by R.A. Sukkar, et al.]

Publication Year: 1992, Page(s):989 - 990
| | PDF (156 KB)

In the above-named work (see ibid., vol.37, no.9, p.1329, 1989), Sukkar et al. defined the function zinc (t)=A sinc ( t)+B cosc (t). The commenter points out that the terms sinc (t) and cosc (t) are forming a pair of Hilbert transforms View full abstract»

• ### On the complex backpropagation algorithm

Publication Year: 1992, Page(s):967 - 969
Cited by:  Papers (148)  |  Patents (1)
| | PDF (224 KB)

A recursive algorithm for updating the coefficients of a neural network structure for complex signals is presented. Various complex activation functions are considered and a practical definition is proposed. The method, associated to a mean-square-error criterion, yields the complex form of the conventional backpropagation algorithm View full abstract»

• ### Fast bit-reversal algorithms based on index representations in GF (2b)

Publication Year: 1992, Page(s):1004 - 1008
Cited by:  Papers (8)
| | PDF (432 KB)

The author proposes bit-reversal unscrambling algorithms based on representing array indices as elements in GF(2b). These elements are sequenced through by counters implemented with integer shifts and bitwise exclusive-OR. A very simple algorithm, developed by applying these counters in a structure similar to the Gold-Rader algorithm, is shown to be less complex and significantly faster... View full abstract»

• ### Nonparametric identification of a particular nonlinear time series system

Publication Year: 1992, Page(s):985 - 989
Cited by:  Papers (11)
| | PDF (364 KB)

A nonlinear time series system is identified. The system has a cascade structure, that is, it consists of a nonlinear memoryless element followed by a dynamic linear system. Given a Gaussian input, a Hermite series based method for recovering the system nonlinearity is presented. The proposed identification procedure is nonparametric since it is able to be consistent for a broad class of nonpolyno... View full abstract»

• ### A filter bank for the directional decomposition of images: theory and design

Publication Year: 1992, Page(s):882 - 893
Cited by:  Papers (470)  |  Patents (2)
| | PDF (1300 KB)

The authors introduce a directionally oriented 2-D filter bank with the property that the individual channels may be critically sampled without loss of information. The passband regions of the component filters are wedge-shaped and thus provide directional information. It is shown that these filter bank outputs may be maximally decimated to achieve a minimum sample representation in a way that per... View full abstract»

• ### Fast algorithms for the discrete Fourier preprocessing transforms

Publication Year: 1992, Page(s):744 - 757
Cited by:  Papers (3)
| | PDF (708 KB)

It is pointed out that the two-stage Fourier algorithms are useful in a variety of applications in signal/image processing and recognition. The first stage, known as the DFPT (discrete Fourier preprocessing transform), has the potential of very fast and low-cost implementation, say, in a VLSI chip, as well as high-quality performance in applications without the second stage. The DFT (discrete Four... View full abstract»

• ### A real-time learning algorithm for a multilayered neural network based on the extended Kalman filter

Publication Year: 1992, Page(s):959 - 966
Cited by:  Papers (137)
| | PDF (564 KB)

A novel real-time learning algorithm for a multilayered neural network is derived from the extended Kalman filter (EKF). Since this EKF-based learning algorithm approximately gives the minimum variance estimate of the linkweights, the convergence performance is improved in comparison with the backwards error propagation algorithm using the steepest descent techniques. Furthermore, tuning parameter... View full abstract»

• ### Robust stability of two-dimensional digital filters under coefficient perturbation

Publication Year: 1992, Page(s):993 - 996
Cited by:  Papers (3)
| | PDF (296 KB)

A bound on allowable weighted additive perturbation in the coefficients of a 2-D stable digital filter from their respective nominal values is determined so that the strict stability of the perturbed-coefficient filter is maintained. The results are suitable for generalization to multidimensional digital filters with real and complex coefficients. Applications of the results may be found in robust... View full abstract»

• ### A new eigenvector weighting method for stable high resolution array processing

Publication Year: 1992, Page(s):999 - 1004
Cited by:  Papers (2)
| | PDF (412 KB)

A stable high resolution estimator (SHIRE) based on a new weighting function for eigenvectors of an array sensor covariance matrix is proposed for bearing estimation of closely spaced sources. SHIRE is data adaptive and requires no user-defined parameter in its implementation. Numerical results are presented to illustrate the performance achievable View full abstract»

• ### A practical method for determining coefficient word length in digital filters

Publication Year: 1992, Page(s):981 - 985
Cited by:  Papers (6)
| | PDF (360 KB)

The author presents a novel method for assessing the sensitivity of recursive digital filters to changes in their coefficients. It is particularly aimed at real-time control applications where continuous filters are being emulated in digital form. The method described enables sensitivity factors to be derived which directly relate the accuracy of the digital filter coefficients to the parameters o... View full abstract»

• ### Modular and numerically stable fast transversal filters for multichannel and multiexperiment RLS

Publication Year: 1992, Page(s):784 - 802
Cited by:  Papers (40)
| | PDF (1484 KB)

The authors present scalar implementations of multichannel and multiexperiment fast recursive least squares algorithms in transversal filter form, known as fast transversal filter (FTF) algorithms. By processing the different channels and/or experiments one at a time, the multichannel and/or multiexperiment algorithm decomposes into a set of intertwined single-channel single-experiment algorithms.... View full abstract»

• ### Multiple invariance ESPRIT

Publication Year: 1992, Page(s):867 - 881
Cited by:  Papers (139)  |  Patents (2)
| | PDF (1260 KB)

A subspace-fitting formulation of the ESPRIT problem is presented that provides a framework for extending the algorithm to exploit arrays with multiple invariances. In particular, a multiple invariance (MI) ESPRIT algorithm is developed and the asymptotic distribution of the estimates is obtained. Simulations are conducted to verify the analysis and to compare the performance of MI ESPRIT with tha... View full abstract»

• ### The quantization effects of the CORDIC algorithm

Publication Year: 1992, Page(s):834 - 844
Cited by:  Papers (117)  |  Patents (5)
| | PDF (736 KB)

A detailed analysis of the quantization error encountered in the CORDIC (coordinate rotation digital computer) algorithm is presented. Two types of quantization error are examined: an approximation error due to the quantized representation of rotation angles, and a rounding error due to the finite precision representation in both fixed-point and floating-point arithmetic. Tight error bounds for th... View full abstract»

• ### Image reconstruction from localized phase

Publication Year: 1992, Page(s):736 - 743
Cited by:  Papers (27)  |  Patents (1)
| | PDF (1100 KB)

The authors present a novel approach to image representation using partial information defined by the localized phase. The scheme is implemented using the short-time (short-distance) Fourier transform. This is a generalization of the Gabor scheme which is well-established with regard to biological representation of visual information at the level of the visual cortex. Similar to processing in visi... View full abstract»

• ### Systolic block Householder transformation for RLS algorithm with two-level pipelined implementation

Publication Year: 1992, Page(s):946 - 958
Cited by:  Papers (36)
| | PDF (876 KB)

The authors propose a systolic block Householder transformation (SBHT) approach to implement the HT on a systolic array and also propose its application to the RLS (recursive least squares) algorithm. Since the data are fetched in a block manner, vector operations are in general required for the vectorized array. However, a modified HT algorithm permits a two-level pipelined implementation of the ... View full abstract»

• ### Combined multiresolution (wide-band/narrow-band) spectrogram

Publication Year: 1992, Page(s):975 - 977
Cited by:  Papers (17)
| | PDF (412 KB)

Existing speech spectrograms-the wideband spectrogram and the narrowband spectrogram-are either deficient in time or frequency resolution. The authors present a method to combine the two spectrograms by evaluating the geometric mean of the corresponding short-time Fourier transform magnitudes. The combined spectrogram preserves the desirable visual features of the originals View full abstract»

• ### CALF: a CORDIC adaptive lattice filter

Publication Year: 1992, Page(s):990 - 993
Cited by:  Papers (17)  |  Patents (1)
| | PDF (280 KB)

CALF is a CORDIC processor implementation of the adaptive lattice filter. Unlike previously reported results which simply use the CORDIC processor to implement the lattice algorithm formulated for conventional multiply-and-accumulate type arithmetic units, a novel adaptive normalized lattice algorithm which directly updates the rotation angles rather than the reflection coefficients is reported. T... View full abstract»

• ### On the application of uniform linear array bearing estimation techniques to uniform circular arrays

Publication Year: 1992, Page(s):1008 - 1011
Cited by:  Papers (30)  |  Patents (2)
| | PDF (320 KB)

The problem of estimating the directions of arrival of narrowband plane waves impinging on a uniform circular array with M identical sensors uniformly distributed around a circle is considered. Specifically, the authors study a transformed data sequence that is equal to a reordering of the inverse discrete Fourier transform sequence corresponding to the outputs of a circular array with View full abstract»

• ### Finite word-length effects in m-D digital filters with singularities on the stability boundary

Publication Year: 1992, Page(s):894 - 900
Cited by:  Papers (17)
| | PDF (452 KB)

The author addresses stability robustness with respect to finite word-length conditions for multidimensional (m-D) digital filters with singularities on the distinguished boundary. In particular, it is shown that all purely first-order digital filters with nonessential singularities of the second kind (NSSKs) are asymptotically stable under certain types of nonlinearities. Even bounded-input, boun... View full abstract»

• ### Regularized fast recursive least squares algorithms for finite memory filtering

Publication Year: 1992, Page(s):758 - 769
Cited by:  Papers (9)
| | PDF (724 KB)

Novel fast recursive least squares algorithms are developed for finite memory filtering, by using a sliding data window. These algorithms allow the use of statistical priors about the solution, and they maintain a balance between a priori and data information. They are well suited for computing a regularized solution, which has better numerical stability properties than the conventional least squa... View full abstract»

• ### A cost minimization approach for optimal window design in spectral analysis of sampled signals

Publication Year: 1992, Page(s):996 - 999
Cited by:  Papers (4)  |  Patents (2)
| | PDF (316 KB)

The problem of digital window design is formulated in terms of cost minimization. This cost quantizes the leakage from one frequency to another. The choice of the penalty function appearing in the cost expression may be adapted to a specific spectral analysis problem and allows the designer to control undesirable leakages. The approach is applied in one particular optimization problem: the minimiz... View full abstract»

• ### The error variance of LMS with time-varying weights

Publication Year: 1992, Page(s):803 - 813
Cited by:  Papers (14)
| | PDF (616 KB)

A calculation of weight error variance of the LMS (least mean square) algorithm is made in the presence of time-varying true weights. With time-varying weights the LMS error system has in general several time scales operating at once. This causes difficulties in the variance calculation which seem hitherto to have passed unnoticed. To handle this problem a sort of perturbation expansion is develop... View full abstract»

• ### Generation and analysis of non-Gaussian Markov time series

Publication Year: 1992, Page(s):845 - 856
Cited by:  Papers (22)
| | PDF (1004 KB)

Correlated non-Gaussian Markov sequences can be considered as filtered white noise (independent, identically distributed sequences of random variables), the filter being a nonlinear system in general. The authors discuss the applicability of linear models and nonlinear methods based on the diagonal series expansion of bivariate densities for analyzing this system. Non-Gaussian sequences exhibit di... View full abstract»

## Aims & Scope

IEEE Transactions on Signal Processing covers novel theory, algorithms, performance analyses and applications of techniques for the processing, understanding, learning, retrieval, mining, and extraction of information from signals

Full Aims & Scope

## Meet Our Editors

Editor-in-Chief
Sergios Theodoridis
University of Athens