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

Acoustics, Speech and Signal Processing, IEEE Transactions on

Issue 1 • Date February 1985

Filter Results

Displaying Results 1 - 25 of 52
  • [Front cover and table of contents]

    Publication Year: 1985 , Page(s): 0
    Save to Project icon | Request Permissions | PDF file iconPDF (314 KB)  
    Freely Available from IEEE
  • VLSI Silicon compilation and the art of automatic microchip design [Book Reviews]

    Publication Year: 1985 , Page(s): 336 - 337
    Save to Project icon | Request Permissions | PDF file iconPDF (297 KB)  
    Freely Available from IEEE
  • Adaptive filtering prediction and control [Book Reviews]

    Publication Year: 1985 , Page(s): 337 - 338
    Save to Project icon | Request Permissions | PDF file iconPDF (318 KB)  
    Freely Available from IEEE
  • LC filter design, test, and manufacturing

    Publication Year: 1985 , Page(s): 337
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (161 KB)  

    First Page of the Article
    View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • [Back cover]

    Publication Year: 1985 , Page(s): c4
    Save to Project icon | Request Permissions | PDF file iconPDF (897 KB)  
    Freely Available from IEEE
  • A wearable tactile intonation display for the deaf

    Publication Year: 1985 , Page(s): 111 - 117
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1232 KB)  

    A wearable device is described which represents the fundamental frequency of voiced sounds as the locus of pitch-synchronous vibrotactile stimulation of the skin. The pitch extractor, which accepts inputs from either a microphone or an accelerometer, uses a combination of low-pass filtering and peak detection to generate a square wave whose frequency is half that of the fundamental frequency of the speech signal. Using a shift register and a clock, the first half of each cycle is timed, the result determining which of eight output channels is actuated during the second half. The output transducer array consists of eight miniature solenoids mounted in a small plastic box. The electronics package is worn on a belt and the solenoid array is mounted on the forearm. The system is powered by thee 9 V NiCad batteries and runs for 2 to 3 h between charges. Experiments with normally hearing subjects confirmed that single channel changes of stimulus location can be detected with relative ease. It was also demonstrated that the system permits discrimination among some of the principal intonation contours of English. The potential value of this device in the rehabilitation of hearing-impaired children is currently under investigation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Texture synthesis using 2-D noncausal autoregressive models

    Publication Year: 1985 , Page(s): 194 - 203
    Cited by:  Papers (74)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1024 KB)  

    The purpose of this paper is to illustrate the usefulness of two dimensional noncausal autoregressive (NCAR) models for the synthesis of textures. These models characterize the gray level at a pixel as a linear combination of gray levels at nearby locations in all directions and an additive white noise variable. We first show that the class of NCAR models is capable of generating a wide variety of image patterns posessing the local replication attribute, an essential ingredient of many natural textures. It is also shown that the theoretical variograms of many NCAR models possess an oscillatory behavior, a characteristic of the variograms of many natural textures. Next, we give experimental results of synthesis of 64 × 64 textures resembling several real textures in the Brodatz album. The synthetic textures generated by 16 parameter NCAR models retain most the visual characteristics of the original textures. The variograms of the original and synthetic textures are also similar. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Sampling expansions for multiband signals

    Publication Year: 1985 , Page(s): 312 - 315
    Cited by:  Papers (8)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (448 KB)  

    The class BIof finite energy signals bandlimited to a set I is considered where I denotes a finite union of disjoint open intervals (bands) on the real frequency axis. If such a signal is sampled uniformly (constant intersample spacing T), it is shown that the following five propositions are equivalent. 1) Any signal in BIis uniquely determined from its uniform samples taken at the rate 1/T samples/s. 2) The set of complex exponentials \{e^{in\omega T}\}_{-\infty}^{\infty} is complete on the set I. 3) the translated sets \{\omega + k\sigma : \omega \epsilon I\} with σ = 2π/T do not intersect I for any k \neq 0 . 4) Each signal in BIpossesses a mean-square convergent sampling expansion in terms of its samples at rate 1/T samples/s. 5) A generalized Parseval relation holds for each pair of signals in BI. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fast algorithms for lpdeconvolution

    Publication Year: 1985 , Page(s): 174 - 182
    Cited by:  Papers (29)  |  Patents (5)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (936 KB)  

    This paper discusses two algorithms for l_{p}, 1 \leq p \leq 2 , deconvolution. One, iterative reweighted least squares (IRLS), is discussed with an eye toward improving its computational efficiency, while the other, residual steepest descent (RSD), is considered in an attempt to improve its convergence properties. In the first case, fast Fourier transforms are used to reduce the number of computations. In the second, a new rescaling procedure which enhances RSD convergence is introduced. The lack of stability of l1, filters, and its implications are discussed. Simple examples are used to illustrate pertinent concepts. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Color display system for connected speech to be used for the hearing impaired

    Publication Year: 1985 , Page(s): 164 - 173
    Cited by:  Papers (4)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1256 KB)  

    A color display system for the hearing impaired which converts connected speech signals into color pictures on a TV screen has been developed. This paper describes the principle, the function and the performance of the system, and the visual images of the patterns displayed using the system. The system consists of a real-time formant tracker, a pitch detector, a memory system, a color coder, etc. In this system, the lowest three formant frequencies are extracted from voiced signals of connected speech by means of the formant tracker, and are converted to three primary color signals. The three primary color signals as a time pattern are represented as a spatial color pattern on the TV screen using the memory system and the color coder. In unvoiced portions, colorless and dapple patterns can be seen. The reproduced Pattern using this system is not only beautiful, but also easy to understand intuitively. Especially, the visual experiments show that simultaneous contrast effect of colors caused by the spatial representation visually compensates for coarticulation effect on connected vowels. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Signal restoration from data aliased in time

    Publication Year: 1985 , Page(s): 151 - 159
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (808 KB)  

    This paper deals with the problem of signal restoration from data aliased in time. The signal, which is in general noncausal, is split into a causal and an anticausal part. The causal part and the time-reversed anticausal part are then modeled as impulse responses of rational pole-zero models. The parameters of these models are then estimated by solving sets of overdetermined equations. The choice of model orders, i.e., number of poles and zeros of rational models, is also discussed. It is shown that if the aliasing period is large enough, there is sufficient information for all the parameters to be estimated. The special cases of a purely causal signal or a purely anticausal signal are discussed. Simulation results show that the signal can be recovered with excellent accuracy. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Ordered and partially ordered processing of multidimensional images

    Publication Year: 1985 , Page(s): 185 - 194
    Cited by:  Papers (6)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (888 KB)  

    Multidimensional images are often converted by scanning to a raster format. For real-time processing a causal property relative to the scan order is important. In other examples the multidimensional image is processed without scanning. In such cases a partially ordered causality property is of interest. The present paper gives a definitive study of the interrelationship between the several causality properties. We establish also a new operator factorization. This factorization can be seen as an extension and refinement of triangular factorization, Shur-Coleski factotization, Gohberg-Krein factorization, and partially ordered factorizations of DeSantis-Porter. An application of the new results to optimal signal extraction and control is summarized. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A class of low-noise computationally efficient recursive digital filters with applications to sampling rate alterations

    Publication Year: 1985 , Page(s): 90 - 97
    Cited by:  Papers (54)  |  Patents (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (776 KB)  

    A new structure for multiband recursive digital filters is proposed. For meeting low-pass filter specifications, it uses fewer multiplications than conventional elliptic filter realizations. An approximation to the minimax solution is obtained numerically by minimizing the LP error norm. The analytic optimum for odd order low-pass filters of this new class turns out to be the elliptic filter itself, but in a new configuration. Analytic solution is also obtained for filters used in decimation/interpolation by a factor of 2. There are several realizations for this new structure, the choice of which depends on the location of poles and zeros. Some selected realizations always have low roundoff noise and small limit cycle bounds. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The radon transforms and some of its applications

    Publication Year: 1985 , Page(s): 338 - 339
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (240 KB)  

    First Page of the Article
    View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Estimates for the stability of low-pass filters

    Publication Year: 1985 , Page(s): 98 - 105
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (656 KB)  

    A measure of stability is defined for digital filters. It is shown that the stability of low-pass filters goes to zero as the filter characteristics approach that of an ideal low-pass filter. More precisely, estimates for the rate of convergence are given. It is shown that, except for constants, the stability of a low-pass filter is no greater than the square root of the reciprocal of the logarithm of the transition bandwidth. For linear phase filters the square root may be removed. These estimates are shown to be optimal for the case of linear phase filters. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Short-time Fourier transform using a bank of low-pass filters

    Publication Year: 1985 , Page(s): 182 - 185
    Cited by:  Papers (1)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (440 KB)  

    The purpose of this paper is to present some experimental results estimating instantaneous frequencies of FM (frequency modulation) signals using the short-time Fourier transform (STFT) method. The relevant STFT coefficients were computed using a bank of simple one-pole low-pass filters which estimates the cross correlation of a given signal and complex sinusoids. Also, it is shown that computing the STFT coefficients is equivalent to finding the least square solution which best approximates the given signal as a weighted sum of complex sinusoids. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Frequency estimation based on iterated autocorrelation function

    Publication Year: 1985 , Page(s): 70 - 76
    Cited by:  Papers (6)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (824 KB)  

    An exclusive time-domain solution has been suggested for estimating the frequency of a single-frequency sinusoid embedded in broad-band noise. The coherent component, viz. the sinusoid is enhanced by repeated or iterated autocorrelation. It is shown that substantial savings in computation can be obtained by coarse quantization of the sequences and then performing the autocorrelations, and that this causes relatively insignificant loss of the usable input SNR range. Various quantization schemes are suggested and the resulting tradeoffs are discussed. Results of a detailed computer simulation experiment conducted to study the performance of the schemes are also given. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimum detection performance of passive coherence estimators

    Publication Year: 1985 , Page(s): 240 - 248
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1080 KB)  

    Two-dimensional coherence processing is commonly used for passive detection in acoustic sensor systems. Modern coherence processors well suited for this purpose include the magnitude-squared coherence (MSC) estimator and the normalized correlation envelope (NCE) estimator. These are biased estimators whose performance characteristics are not well understood and are exceptionally difficult to analyze. However, by developing a rather unique bias-correcting function, the detection performance of biased estimators is derived as a function of both the input-signal and the processor parameters. Design parameters which optimize detection are determined for both the MSC and the NCE estimators. In applications where the number of degrees of freedom (or time-bandwidth product) of a target signal is severely limited, detection performance can be enhanced by signal overcontainment within the processor bandwidth. Performance data are presented in formats appropriate for selecting processor parameters which are optimum for a given application. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimal solution of a training problem in speech recognition

    Publication Year: 1985 , Page(s): 326 - 329
    Cited by:  Papers (16)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (512 KB)  

    We take the view that the payoff correpsonding to different ways of training a speech recognizer is the probability that the recognizer will correctly recognize a randomly chosen word. In "A Decision Theoretic Formulation of a Training Problem in Speech Recognition" we posed the problem of choosing a speech recognizer as an optimization problem with the expected value of the above payoff as the objective function. This correspondence presents the optimal Bayes solution to this optimization problem by maximizing the expected payoff: conditionally on given training data decode the acoustic signal for a word as any word which maximizes the a posteriori expected joint probability of the word and the acoustic signal. Thus the probability estimator which is optimal for mean-squared error produces a decoder which happens to be optimal for recognition rate as well. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The use of multiple criterion optimization for frequency domain design of noncausal IIR filters

    Publication Year: 1985 , Page(s): 126 - 135
    Cited by:  Papers (7)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (840 KB)  

    Noncausal recursive filters have recently been proposed because of their increased efficiency versus linear phase FIR filters and causal IIR filters [2]. A general frequency domain design technique capable of exploiting the characteristics of noncausal filters is not yet available. This paper proposes the use of multiple criterion optimization for the frequency domain design of noncausal IIR filters. A new family of filters is generated and their efficiency versus causal FIR filters and IIR filters is discussed. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Variance and bias of the normalized relay correlation estimator

    Publication Year: 1985 , Page(s): 334 - 336
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (392 KB)  

    In a recent paper, Gabriel states that no simple or enlightening expression exists for the variance and bias of the normalized relay correlation estimator for independent pairs of Gaussian samples. In this note, approximate expressions for this variance and bias are derived which are both simple and agree with the empirical variances measured by Gabriel. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Positive time-frequency distribution functions

    Publication Year: 1985 , Page(s): 31 - 38
    Cited by:  Papers (74)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (840 KB)  

    We demonstrate the existence of positive joint distributions of time and frequency for arbitrary signals. A method is given to readily generate an infinite number of them for any signal. General properties of these distribution functions are derived and specific examples for some common signals are presented. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Root properties and convergence rates of median filters

    Publication Year: 1985 , Page(s): 230 - 240
    Cited by:  Papers (43)  |  Patents (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1120 KB)  

    Median filters are a special class of ranked order filters used for smoothing signals. Repeated application of the filter on a quantized signal of finite length ultimately results in a sequence, termed a root signal, which is invariant to additional passes of the median filter. In this paper, the theory is developed both for determining the cardinality of the root signal space of arbitrary window width filters applied to signals with any number of quantization levels and for counting or estimating the number of passes required to produce a root for binary signals. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Two-dimensional linear prediction models--part I: Spectral factorization and realization

    Publication Year: 1985 , Page(s): 280 - 299
    Cited by:  Papers (23)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1752 KB)  

    In this paper we present several results for three different canonical forms of linear prediction on a plane. These filters have causal, semicausal, and noncausal prediction geometries. Starting from their properties we consider the problem of realization of these filters from a given power spectral density function (SDF). Since it is not possible in general to obtain rational spectral factors of a two-dimensional SDF, we propose algorithms for obtaining rational approximations which are stable and converge to their limit (irrational) factors as the order of approximation is increased. It is also shown that the normal equations associated with the minimum variance two-dimensional prediction filters give a useful algorithm for obtaining rational approximations which are stable and converge to their unique limit filters. This result allows design of finite-order stable filters by solving a finite number of equations while realizing the given SDF arbitrarily closely. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Time correlation statistics of the LMS adaptive algorithm weights

    Publication Year: 1985 , Page(s): 309 - 312
    Cited by:  Papers (10)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (488 KB)  

    The transient and steady-state weight correlation statistics of both the real and complex LMS adaptive filters are obtained when the inputs are independent samples from real and circularly Gaussian processes, respectively. A matrix relationship is derived between the covariance matrix of the weight vector at two different times and the covariance matrix of the weights at one time. These expressions show that the weight fluctuations have the same time constants as the mean behavior of the LMS algorithm itself (i.e., the weights are correlated over the same number of iterations that it takes for the algorithm to converge to the Wiener weights for stationary inputs). View full abstract»

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

Aims & Scope

This Transactions ceased production in 1990. The current retitled publication is IEEE Transactions on Signal Processing.

Full Aims & Scope