IEEE Transactions on Signal Processing

Issue 3 • March 1991

Filter Results

Displaying Results 1 - 25 of 33
  • Corrections to 'Gradient algorithms for designing predictive vector quantizers'

    Publication Year: 1991, Page(s):764 - 765
    Cited by:  Papers (9)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (179 KB)

    An error in a paper by Chang and Gray (see ibid., vol.34, no.4, pp. 679-90, 1986) is pointed out and corrected. The error invalidates their observation that the generalized Lloyd algorithm is a gradient descent technique but the generalized Lloyd algorithm is a member of the related class of coordinate descent techniques. Convergence rate analysis of gradient descent algorithms for vector quantize... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Factorization of the coefficient variance matrix in orthogonal transforms

    Publication Year: 1991, Page(s):714 - 718
    Cited by:  Papers (6)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (392 KB)

    The authors derive a matrix factorization of the coefficient variances for any fixed transformation. This factorization formulation provides a better framework for understanding the role of transform on the one hand, and the signal source model on the other. It also greatly simplifies the number of computational operations required for each variance calculation. A transform-specific matrix which l... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Connected-digit speaker-dependent speech recognition using a neural network with time-delayed connections

    Publication Year: 1991, Page(s):698 - 713
    Cited by:  Papers (44)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (1396 KB)

    An analog neural network that can be taught to recognize stimulus sequences is used to recognize the digits in connected speech. The circuit computes in the analog domain, using linear circuits for signal filtering and nonlinear circuits for simple decisions, feature extraction, and noise suppression. An analog perceptron learning rule is used to organize the subset of connections used in the circ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • High resolution two-dimensional ARMA spectral estimation

    Publication Year: 1991, Page(s):765 - 770
    Cited by:  Papers (23)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (452 KB)

    The authors present a practical algorithm for estimating the power spectrum of a 2-D homogeneous random field based on 2-D autoregressive moving average (ARMA) modeling. This algorithm is a two-step approach: first, the AR parameters are estimated by solving a version of the 2-D modified Yule-Walker equation, for which some existing efficient algorithms are available; then the MA spectrum paramete... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A class of fast Gaussian binomial filters for speech and image processing

    Publication Year: 1991, Page(s):723 - 727
    Cited by:  Papers (53)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (380 KB)

    The authors present an efficient, in-place algorithm for the batch processing of linear data arrays. These algorithms are efficient, easily scaled, and have no multiply operations. They are suitable as front-end filters for a bank of quadrature mirror filters and for pyramid coding of images. In the latter application, the binomial filter was used as the low-pass filter in pyramid coding of images... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Compound Gauss-Markov random fields for image estimation

    Publication Year: 1991, Page(s):683 - 697
    Cited by:  Papers (126)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (3752 KB)

    Algorithms for obtaining approximations to statistically optimal estimates for images modeled as compound Gauss-Markov random fields are discussed. The authors consider both the maximum a posteriori probability (MAP) estimate and the minimum mean-squared error (MMSE) estimate for image estimation and image restoration. Compound image models consist of several submodels having different characteris... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A time-domain interpretation of the KL spectrum

    Publication Year: 1991, Page(s):721 - 723
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (272 KB)

    An information theoretic spectrum minimizing the frequency-domain Kullback-Leibler (KL) information number has been proposed by B.S. Choi (IEEE Trans. Acoust., Speech, and Signal Processing, vol.38, p.717-21, 1990). Based on the KL spectrum, a method of modifying a spectrum estimate was suggested. Some numerical examples illustrated that the KL spectrum estimate is superior to the initial estimate... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Maximum likelihood estimation of object location in diffraction tomography

    Publication Year: 1991, Page(s):672 - 682
    Cited by:  Papers (33)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (988 KB)

    The problem is formulated within the context of diffraction tomography, where the complex phase of the diffracted wavefield is modeled using the Rytov approximation and the measurements consist of noisy renditions of this complex phase at a single frequency. The log likelihood function is computed for the case of additive zero mean Gaussian white noise and shown to be expressible in the form of th... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • New algorithm for solving block matrix equations with applications in 2-D AR spectral estimation

    Publication Year: 1991, Page(s):759 - 764
    Cited by:  Papers (13)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (416 KB)

    The author presents an iterative algorithm to solve Toeplitz and non-Toeplitz block matrix equations. The development is based upon some well-known matrix iterative techniques. The algorithm is developed for the ideal case where the individual blocks in the autocorrelation matrix are Toeplitz, and it is then extended to a more general least squares data formulation case. It is shown that the algor... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive recovery of a chirped sinusoid in noise. I. Performance of the RLS algorithm

    Publication Year: 1991, Page(s):583 - 594
    Cited by:  Papers (69)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (908 KB)

    The authors study the ability of the exponentially weighted recursive least square (RLS) algorithm to track a complex chirped exponential signal buried in additive white Gaussian noise (power P n). The signal is a sinusoid whose frequency is drifting at a constant rate Ψ. lt is recovered using an M-tap adaptive predictor. Five principal aspects of the study are pre... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On the family of ML spectral estimates for mixed spectrum identification

    Publication Year: 1991, Page(s):644 - 655
    Cited by:  Papers (29)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (1000 KB)

    A recently developed point spectrum identification procedure based on a family of AR and ML spectral estimates is exploited to arrive at a mixed spectrum identification procedure. To this end, a variety of properties of the AR and ML estimates as a function of model order are described. These properties relate to amplitude convergence, resolution and a characterization of the AR spectral artifact ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Instantaneous frequency and conditional moments in the time-frequency plane

    Publication Year: 1991, Page(s):755 - 756
    Cited by:  Papers (8)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (160 KB)

    The conditional moments of frequency of a time-frequency distribution are discussed, and it is shown that they may be related to the derivative of the log of the corresponding signal, a complex function whose imaginary part is the instantaneous frequency. For the complex energy spectrum, a general form for the moments is obtained. For a signal which is the sum of several other signals, the instant... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On the application of mixture AR hidden Markov models to text independent speaker recognition

    Publication Year: 1991, Page(s):563 - 570
    Cited by:  Papers (45)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (664 KB)

    Linear predictive hidden Markov models have proved to be efficient for statistically modeling speech signals. The possible application of such models to statistical characterization of the speaker himself is described and evaluated. The results show that even with a short sequence of only four isolated digits, a speaker can be verified with an average equal-error rate of less than 3 %. These resul... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • FIR prefiltering improves Prony's method

    Publication Year: 1991, Page(s):736 - 741
    Cited by:  Papers (17)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (476 KB)

    Prony's method, a technique for determining the parameters of sinusoidal signals, is known to be very sensitive to measurement errors in the observed signal samples. The authors present two prefiltering methods that improve the performance of Prony's method. The simple prefiltering operation reduces the effects of noise in the data without violating the fundamental property of Prony's method. In t... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On the peaks of causal signals with a given average delay

    Publication Year: 1991, Page(s):620 - 626
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (512 KB)

    Bounds are derived on the maximum location and minimum amplitude of the peak of a causal signal with a given average delay. For an average delay of τ, the authors show that the maximum possible location of the signal peak is on the order of τ(τ+3)/2. This bound can also be interpreted as providing the maximum integer at which the most probable value of a discrete nonnegative random var... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The Cramer-Rao lower bound for signals with constant amplitude and polynomial phase

    Publication Year: 1991, Page(s):749 - 752
    Cited by:  Papers (161)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (304 KB)

    The authors derive the Cramer-Rao lower bound (CRLB) for complex signals with constant amplitude and polynomial phase, measured in additive Gaussian white noise. The exact bound requires numerical inversion of an ill-conditioned matrix, while its O(N -1) approximation is free of matrix inversion. The approximation is tested for several typical parameter values and is f... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An experimental demonstration of improved Doppler processing performance

    Publication Year: 1991, Page(s):718 - 721
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (348 KB)

    The authors present the experimental results of applying the wideband modified forward-backward linear prediction (MFBLP) algorithm to radar data obtained in a real target-clutter environment for detection improvement in the Doppler frequency domain. The system used for data acquisition is an L-band radar. The system parameters and specifications relevant to this experiment are given. Thi... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Nonparametric bispectrum-based time-delay estimators for multiple sensor data

    Publication Year: 1991, Page(s):770 - 774
    Cited by:  Papers (12)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (316 KB)

    When estimating time-delays of signals from sensor outputs, the bispectrum has advantages when the signal has a skewed distribution. The authors provide nonparametric estimation approaches which exploit the two-dimensional nature of the bispectrum. When dealing with observations from more than a pair of sensors, the new algorithms have computational advantages and, in some cases, show improved per... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Computing time-frequency distributions [signal analysis]

    Publication Year: 1991, Page(s):727 - 729
    Cited by:  Papers (4)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (280 KB)

    Recently, numerous strategies have been proposed for computing discrete time-frequency distributions such as the Wigner distribution. The author describes an efficient and straightforward strategy for computing time-frequency distributions that are members of Cohen's class. The strategy is based on the work by A.H. Nuttall (Naval Underwater Syst. Center, Tech. Rep., 8533, April 1989) which has res... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive recovery of a chirped sinusoid in noise. II. Performance of the LMS algorithm

    Publication Year: 1991, Page(s):595 - 602
    Cited by:  Papers (43)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (580 KB)

    For pt.I see ibid., vol.39, no. 3, p.583-94 (1991). The authors present a methodology for evaluating the tracking behavior of the least-mean square (LMS) algorithm for the nontrivial case of recovering a chirped sinusoid in additive noise. A complete closed-form analysis of the LMS tracking properties for a nonstationary inverse system modeling problem is also presented. The mean-square error (MSE... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Maximum likelihood based sensor array signal processing in the beamspace domain for low angle radar tracking

    Publication Year: 1991, Page(s):656 - 671
    Cited by:  Papers (29)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (1292 KB)

    The novelty of the 3D beamspace domain, maximum likelihood (3D BDML) bearing estimation scheme is its judicious exploitation of the fact that the respective beams associated with any three classical beam-forming vectors which are mutually orthogonal have M-3 nulls in common, where M is the number of elements comprising a uniformly spaced, linear array. Exploitation of this proper... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • DPCM coding of spectral amplitudes without positive slope overload

    Publication Year: 1991, Page(s):756 - 758
    Cited by:  Patents (3)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (252 KB)

    The author considers the problem of coding the spectral amplitudes of a sinusoidal transform coding (STC) vocoder operating at 2400 b/s, which was previously considered by R.J. McAulay and T.F. Quatieri (1987). Differential pulse code modulation (DPCM) coding of spectral amplitudes in vocoders can cause formant clipping because of positive slope overload. The author presents a modified DPCM coder ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Efficient adaptive algorithms for ARX identification

    Publication Year: 1991, Page(s):571 - 582
    Cited by:  Papers (7)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (696 KB)

    Two efficient adaptive algorithmic families are developed for multichannel combiners characterized by unequal memory lengths for each input channel. The prewindowed, the covariance, and the sliding window case are addressed. The difference between the proposed methods lies in the way the Kalman gain vector is order-updated in each case. The first algorithm operates on both inputs simultaneously an... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new technique for velocity estimation of large moving objects

    Publication Year: 1991, Page(s):741 - 743
    Cited by:  Papers (6)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (284 KB)

    A technique for motion estimation of large moving objects is presented. This technique is based on analyzing the Hartley transform spectrum of the image sequence directly, instead of using it to compute other transforms. In this technique, the fast Hartley transform (FHT) is applied to the image sequence and followed by a peak detection procedure. The location of the peak is related to the velocit... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A robust approach for high resolution frequency estimation

    Publication Year: 1991, Page(s):627 - 643
    Cited by:  Papers (20)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (1068 KB)

    A robust estimation method for frequencies of received signals is considered. The influence function of a robust estimate is derived for the sinusoidal signals. The variance of the robust estimate is also derived using the influence function. It is found that the robust estimate attains the Cramer-Rao lower bound (CRLB) for the contaminated Gaussian distribution. It is of order O(N View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.

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
Pier Luigi Dragotti
Imperial College London