Scheduled System Maintenance:
On Monday, April 27th, IEEE Xplore will undergo scheduled maintenance from 1:00 PM - 3:00 PM ET (17:00 - 19:00 UTC). No interruption in service is anticipated.
By Topic

Signal Processing, IEEE Transactions on

Issue 5 • Date May 1999

Filter Results

Displaying Results 1 - 25 of 40
  • Comments on "On a least-squares-based algorithm for identification of stochastic linear systems"

    Publication Year: 1999 , Page(s): 1395 - 1396
    Cited by:  Papers (13)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (74 KB)  

    The bias-eliminated least squares estimator proposed in the paper by Zheng (see ibid., vol.46, p.1631-8, 1998) is shown to be identical to a simple instrumental variable estimator, using delayed input values as instruments. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An overview of aliasing errors in discrete-time formulations of time-frequency representations

    Publication Year: 1999 , Page(s): 1463 - 1474
    Cited by:  Papers (11)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (572 KB)  

    Discrete-time time-frequency representation (TFR) algorithms claim to provide alias-free approximations to their continuous-time TFR counterparts without requiring oversampling of the signal By counterexamples, we demonstrate that some of these claims are invalid. We give new necessary conditions for reducing aliasing errors in these discrete-time TFR algorithms View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Localization of wideband signals using least-squares and total least-squares approaches

    Publication Year: 1999 , Page(s): 1213 - 1222
    Cited by:  Papers (19)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (492 KB)  

    In this paper, we introduce a new focusing technique for localization of wideband signals. Relaxing the unitary assumption for the focusing matrices, we formulate the least-square (LS) and the total least-square (TLS) coherent signal-subspace methods. The TLS is an alternative to the conventional LS and uses the fact that errors can exist both in the focusing location matrix as well as in the estimated location matrix at a given frequency bin. To prevent the focusing loss, we use a class of focusing matrices that are constant under multiplication by their Hermitian transpose. The class of unitary matrices comports with this property. We then develop a new focusing technique based on a modification to the TLS (MTLS). It is shown that the computational complexity of the new technique is significantly lower than that for the rotational signal subspace method (RSS). The focusing gain of the new technique is also larger than the focusing gain of the RSS algorithm. The simulation study shows that, compared with the RSS, the new algorithm has a smaller resolution signal to-noise ratio (SNR) View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new method for D-dimensional exact deconvolution

    Publication Year: 1999 , Page(s): 1324 - 1334
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (708 KB)  

    Deconvolution is an important problem of signal processing, and conventional approaches, including Fourier methods, have stability problems due to the zeros of the convolution kernel. We present a new method of multidimensional exact deconvolution. This method is always stable, even when the convolution kernel h(n) has zeros on the unit circle, and there exist closed-form solutions for the one-dimensional (1-D) case (D=1). For the multidimensional case (D>1), the proposed method yields stable solutions when det(h)=D. This solution set covers a portion of all possible convolution kernels, including the ones that have zeros on the multidimensional unit circle. This novel time-domain method is based on the fact that the convolution inverse of a first-order kernel can be found exactly in multidimensional space. Convolution inverses for higher order kernels are obtained using this fact and the zeros of the convolution kernel. The presented method is exact, stable, and computationally efficient. Several examples are given in order to show the performance of this method in 1-D and multidimensional cases View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A modified path-following algorithm using a known algebraic path

    Publication Year: 1999 , Page(s): 1407 - 1409
    Cited by:  Papers (7)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (120 KB)  

    A new computationally efficient path-following algorithm is presented for source localization that uses a uniform linear sensor array. An algebraic relation is demonstrated between the bearing of a source under a plane wave assumption and the actual source bearing and range. This relation can be used as a path to follow to the peak of the 2-D MUSIC spectrum. As a result, in the case of an array that has m sensors with n sources, this new algorithm can reduce the number of searches to n independent 1-D searches following the known algebraic path View full abstract»

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

    Publication Year: 1999 , Page(s): 1304 - 1313
    Cited by:  Papers (24)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (304 KB)  

    The discrete wavelet transform (DWT) is usually carried out by filterbank iteration; however, for a fixed number of zero moments, this does not yield a discrete-time basis that is optimal with respect to time localization. This paper discusses the implementation and properties of an orthogonal DWT, with two zero moments and with improved time localization. The basis is not based on filterbank iteration; instead, different filters are used for each scale. For coarse scales, the support of the discrete-time basis functions approaches two thirds that of the corresponding functions obtained by filterbank iteration. This basis, which is a special case of a class of bases described by Alpert (1992, 1993), retains the octave-band characteristic and is piecewise linear (but discontinuous). Closed-form expressions for the filters are given, an efficient implementation of the transform is described, and improvement in a denoising example is shown. This basis, being piecewise linear, is reminiscent of the slant transform, to which it is compared View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Complexity reduction of the NLMS algorithm via selective coefficient update

    Publication Year: 1999 , Page(s): 1421 - 1424
    Cited by:  Papers (55)  |  Patents (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (180 KB)  

    This article proposes an algorithm for partial update of the coefficients of the normalized least mean square (NLMS) finite impulse response (FIR) adaptive filter. It is shown that while the proposed algorithm reduces the complexity of the adaptive filter, it maintains the closest performance to the full update NLMS filter for a given number of updates. Analysis of the MSE convergence and steady-state performance for independent and identically distributed (i.i.d.) signals is provided for the extreme case of one update/iteration View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A general efficient method for chaotic signal estimation

    Publication Year: 1999 , Page(s): 1424 - 1428
    Cited by:  Papers (30)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (204 KB)  

    This article presents a general, computationally efficient method for chaotic signal estimation based on the connection between the symbolic sequence and the initial condition of a chaotic system. The performance of the method in white Gaussian noise is evaluated. The new method is asymptotically unbiased and attains the Cramer-Rao lower bound at high signal-to-noise ratios View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Two-dimensional FIR filter design by transformation

    Publication Year: 1999 , Page(s): 1474 - 1478
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (236 KB)  

    This article extends the McClellan (1973) transformation method so that new types of one-dimensional (1-D) filters can be transformed, and new types of two-dimensional (2-D) filters can be designed. Two procedures are presented for the design of complex and real, positive- and negative-symmetric, 2-D filters. The first procedure is used to design complex 2-D filters with a rectangular region of support having odd-length sides, whereas the second procedure is used for filters with a rectangular region of support having even-length sides View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Maximum-likelihood symmetric α-stable parameter estimation

    Publication Year: 1999 , Page(s): 1382 - 1384
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (112 KB)  

    Using the close relation between Fisher scoring and Newton maximization, and an efficient density function evaluation, we develop a fast maximum-likelihood parameter estimation method. Simulations show the algorithm to be superior in accuracy to McCulloch's (1986) method and to achieve the Cramer-Rao bound View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A modified QRD for smoothing and a QRD-LSL smoothing algorithm

    Publication Year: 1999 , Page(s): 1414 - 1420
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (256 KB)  

    This paper introduces a modified QR-decomposition (QRD) that extends the method of QRD to a more general case to solve the least-squares lattice smoothing problems. We show that the conventional QRD is a special form of the modified QRD that occurs when no future data values are used. Within the framework of the modified QRD procedure, an order-recursive QRD based least-squares lattice (QRD-LSL) smoothing algorithm is formulated. The algorithm combines all the desirable features of the standard QRD-LSL filtering algorithm with a more accurate smoothing process. The results of some computer simulations of a channel equalizer are also presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The probability density of spectral estimates based on modified periodogram averages

    Publication Year: 1999 , Page(s): 1255 - 1261
    Cited by:  Papers (20)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (284 KB)  

    Welch's (1967) method for spectral estimation of averaging modified periodograms has been widely used for decades. Because such an estimate relies on random data, the estimate is also a random variable with some probability density function. Here, the PDF of a power estimate is derived for an estimate based on an arbitrary number of frequency bins, overlapping data segments, amount of overlap, and type of data window, given a correlated Gaussian input sequence. The PDFs of several cases are plotted and found to be distinctly non-Gaussian (the asymptotic result of averaging frequency bins and/or data segments), using the Kullback-Leibler distance as a measure. For limited numbers of frequency bins or data segments, the precise PDF is considerably skewed and will be important in applications such as maximum likelihood tests View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On adaptive EVD asymptotic distribution of centro-symmetric covariance matrices

    Publication Year: 1999 , Page(s): 1402 - 1406
    Cited by:  Papers (7)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (272 KB)  

    This article investigates the gain in statistical performance/complexity of the adaptive estimation of the eigenvalue decomposition (EVD) of covariance matrices when the centro-symmetric (CS) structure of such matrices is utilized. After deriving the asymptotic distribution of the EVD estimators, it is shown, in particular, that the closed-form expressions for the asymptotic covariance of batch and adaptive EVD estimators are very similar, provided that the number of samples is replaced by the inverse of the step size View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Computationally efficient maximum likelihood estimation of structured covariance matrices

    Publication Year: 1999 , Page(s): 1314 - 1323
    Cited by:  Papers (35)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (352 KB)  

    By invoking the extended invariance principle (EXIP), we present herein a computationally efficient method that provides asymptotic (for large samples) maximum likelihood (AML) estimation for structured covariance matrices and is referred to as the AML algorithm. A closed-form formula for estimating the Hermitian Toeplitz covariance matrices that makes AML computationally simpler than most existing Hermitian Toeplitz matrix estimation algorithms is derived. Although the AML covariance matrix estimator can be used in a variety of applications, we focus on array processing. Our simulation study shows that AML enhances the performance of angle estimation algorithms, such as MUSIC, by making them very close to the corresponding Cramer-Rao bound (CRB) for uncorrelated signals. Numerical comparisons with several structured and unstructured covariance matrix estimators are also presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Estimation of fractional Brownian motion with multiresolution Kalman filter banks

    Publication Year: 1999 , Page(s): 1431 - 1434
    Cited by:  Papers (4)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (140 KB)  

    A filter bank design based on orthonormal wavelets and equipped with a multiscale Kalman filter was proposed for deconvolution of fractal signals. We use the same scheme for estimating fractional Brownian motion in noise considering (1) the effect of correlation in the sequence of wavelet coefficients; (2) the approximation term in the wavelet expansion; (3) aliasing effects; (4) the optimal number of scales in the filter bank. Considerations on the minimum number of filters in the bank are made, and comparisons between Wiener and Kalman filters are given. Explicit expressions of the mean-square error are given, and comparisons between theoretical and simulation results are shown View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Estimation of continuous-time AR process parameters from discrete-time data

    Publication Year: 1999 , Page(s): 1232 - 1244
    Cited by:  Papers (25)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (544 KB)  

    The problem of estimating continuous-time autoregressive process parameters from discrete-time data is considered. The basic approach used here is based on replacing the derivatives in the model by discrete-time differences, forming a linear regression, and using the least squares method. Such a procedure is simple to apply, computationally flexible and efficient, and may have good numerical properties. It is known, however, that all standard approximations of the highest order derivative, such as repeated use of the delta operator, gives a biased least squares estimate, even as the sampling interval tends to zero. Some of our previous approaches to overcome this problem are reviewed. Then. two new methods, which avoid the shift in our previous results, are presented. One of them, which is termed bias compensation, is computationally very efficient. Finally, the relationship of the above least squares approaches with an instrumental variable method is investigated. Comparative simulation results are also presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Methods for blind equalization and resolution of overlapping echoes of unknown shape

    Publication Year: 1999 , Page(s): 1245 - 1254
    Cited by:  Papers (16)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (300 KB)  

    This paper considers the related problems of using an uncalibrated antenna array to (1) recover an unknown signal transmitted over an unknown (but stationary) multipath channel and (2) resolve overlapping pulse echoes with unknown shape. Unlike previously proposed multichannel blind equalization techniques, the methods described herein employ a model based on physical channel parameters rather than unstructured single-input, multi-output FIR filters. The algorithms exploit similarities between a model for the data in the frequency domain and the standard direction-of-arrival estimation problem. This connection between the two problems suggests several different approaches based on, for example, maximum likelihood, MODE, IQML, and ESPRIT. These approaches are developed in some detail, and the results of several simulation examples are included to compare their performance View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Equivariant adaptive selective transmission

    Publication Year: 1999 , Page(s): 1223 - 1231
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (368 KB)  

    In this paper, we consider the problem of selective transmission-the dual of the blind source separation task-in which a set of independent source signals are adaptively premixed prior to a nondispersive physical mixing process so that each source can be independently monitored in the far field. Following similar procedures for information-theoretic blind source separation, we derive a stochastic gradient algorithm for iteratively estimating the premixing matrix in the selective transmission problem, and through a simple modification, we obtain a second algorithm whose performance is equivariant with respect to the channel's mixing characteristics. The local stability conditions for the algorithms about any selective transmission solution are shown to be the same as those for similar source separation algorithms. Practical implementation issues are discussed, including the estimation of the combined system matrix and the reordering and scaling of the received signals within the algorithm. Mean square error-based selective transmission algorithms are also derived for performance comparison purposes. Simulations indicate the useful behavior of the premixing algorithms for selective transmission View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On the recursive solution of the normal equations of bilateral multivariate autoregressive models

    Publication Year: 1999 , Page(s): 1388 - 1390
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (120 KB)  

    A multivariate version of the bilateral autoregressive (AR) model is proposed, and a recursive algorithm is presented to solve the normal equations of the bilateral multivariate AR models. The recursive algorithm is computationally efficient and easy to implement as a computer program. The recursive algorithm is useful for identifying and smoothing not only bilateral multivariate AR processes but multidimensional multivariate AR processes and multivariate spatio-temporal processes as well View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Discrete fractional Fourier transform based on orthogonal projections

    Publication Year: 1999 , Page(s): 1335 - 1348
    Cited by:  Papers (107)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (580 KB)  

    The continuous fractional Fourier transform (FRFT) performs a spectrum rotation of signal in the time-frequency plane, and it becomes an important tool for time-varying signal analysis. A discrete fractional Fourier transform has been developed by Santhanam and McClellan (see ibid., vol.42, p.994-98, 1996) but its results do not match those of the corresponding continuous fractional Fourier transforms. We propose a new discrete fractional Fourier transform (DFRFT). The new DFRFT has DFT Hermite eigenvectors and retains the eigenvalue-eigenfunction relation as a continous FRFT. To obtain DFT Hermite eigenvectors, two orthogonal projection methods are introduced. Thus, the new DFRFT will provide similar transform and rotational properties as those of continuous fractional Fourier transforms. Moreover, the relationship between FRFT and the proposed DFRFT has been established in the same way as the conventional DFT-to-continuous-Fourier transform View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Blind adaptive FRESH filtering for signal extraction

    Publication Year: 1999 , Page(s): 1397 - 1402
    Cited by:  Papers (14)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (256 KB)  

    A blind adaptive method called the BA-FRESH (optimum frequency-shift) filtering technique is proposed and its convergence analyzed. The BA-FRESH filter does not require training signals nor knowledge of the statistics of the desired signal. It is capable of separating desired signals from spectrally overlapping interference by knowing only the cycle frequencies of the signals View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multicomponent polynomial phase signal analysis using a tracking algorithm

    Publication Year: 1999 , Page(s): 1390 - 1395
    Cited by:  Papers (5)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (220 KB)  

    We describe an efficient technique analyzing signals that comprise a number of polynomial phase components. The technique is based on a previously proposed “multiple frequency tracker”, which is an algorithm for recursive estimation of parameters of multiple sine waves in noise. It has a relatively low SNR threshold and moderate computational complexity View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fuzzy selection filters for image restoration with neural learning

    Publication Year: 1999 , Page(s): 1446 - 1450
    Cited by:  Papers (5)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (180 KB)  

    A novel class of nonlinear filters, called rank conditioned fuzzy selection (RCFS) filters, is proposed to improve the filtering capability of rank conditioned rank selection filters. In contrast to the selection filters, the output of RCFS filters is obtained from the center gravity of a selected fuzzy set of the observation samples View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A two-level interleaving architecture for serial convolvers

    Publication Year: 1999 , Page(s): 1481 - 1486
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (280 KB)  

    We present a bit-serial architecture for convolving/correlating long numerical sequences by long filter functions. Because of its two-level interleaving structure, the proposed device does not require “wait cycles” between consecutive input samples. As a result, it achieves the highest possible throughput. Cascadability, fault tolerance, feasibility in VLSI technology, and computing performances are discussed and analyzed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multichannel system identification and deconvolution: performance bounds

    Publication Year: 1999 , Page(s): 1410 - 1414
    Cited by:  Papers (11)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (200 KB)  

    We consider the problem of estimating the parameters of an unknown multi-input multi-output (MIMO) linear system and the related problem of deconvolving and recovering its inputs. Only the system outputs are assumed to be observable. The system inputs are assumed to be non-Gaussian. We derive simple closed-form asymptotic expressions for the Cramer-Rao lower bound (CRLB) for the system parameters, as well as lower bounds on the signal reconstruction performance. These show that the identification/deconvolution performance depend on the accuracy with which the location (mean) and the scale (standard deviation) parameters of the input probability density functions can be identified from observation of the input signals 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
Sergios Theodoridis
University of Athens