By Topic

Acoustics, Speech and Signal Processing, IEEE Transactions on

Issue 2 • Date Feb 1989

Filter Results

Displaying Results 1 - 16 of 16
  • L1 and L minimization via a variant of Karmarkar's algorithm

    Page(s): 245 - 253
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (688 KB)  

    Simple iterative algorithms are presented for L1 and L minimization (regression) based on a variant of Karmarkar's linear programming algorithm. Although these algorithms are based on entirely different theoretical principles to the popular IRLS (iteratively reweighted least squares) algorithm, they have almost identical matrix operations. Also presented are the results of a Monte Carlo study comparing the numerical convergence properties of the Karmarkar algorithm for L1 minimization to those of an IRLS and a simplex algorithm. The test problem involves L 1 estimation of AR (autoregressive) model parameters. The Karmarkar algorithm outperformed IRLS by achieving higher numerical accuracy in fewer iterations. Techniques for reducing the computational cost per iteration of the Karmarkar L1 algorithm are discussed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Improved addition for the logarithmic number system

    Page(s): 301 - 303
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (280 KB)  

    The logarithmic number system (LNS), which offers a wide dynamic range with an independently choosable signal-to-noise ratio, is discussed. To realize addition, voluminous lookup tables are needed. A method is proposed to reduce the storage requirements of these tables by means of Chebyshev approximation with unequally spaced partition points View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Analysis of an adaptive technique for modeling sparse systems

    Page(s): 254 - 264
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (764 KB)  

    An effective technique for modeling sparse systems has been developed, and the error surface of the system is analyzed with respect to estimation noise. The technique requires a type of adaptive filter which is called an adaptive delay filter. An implementation of the adaptive delay filter is discussed that includes adaptive gains in addition to variable delay taps. The filter is especially applicable to modeling systems with a sparse impulse response. Less computation is required for a sparse system than with the conventional approach. The technique is tested with a variety of unknown systems using both white noise input and autoregressive input. It is shown that it works properly for both sparse and nonsparse systems in noise-free and noisy conditions. The performance of the technique is verified by a careful analysis of the error surface and the techniques for delay determination and corresponding gain adaption View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Frequency difference of arrival accuracy

    Page(s): 306 - 308
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (216 KB)  

    The problem considered occurs in the geolocation of a fixed emitter using observations from two moving collectors. Estimates for the time difference and frequency difference of signal arrivals are used. Frequency difference of arrival (FDOA) is estimated through the use of a mixing product. Standard regression analysis procedures are then applied to estimate the slope of the unwrapped phase angle. The derivation of an expression for the standard error of FDOA is given, and this result is related to several other well-known results View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Maximum likelihood noise cancellation using the EM algorithm

    Page(s): 204 - 216
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (940 KB)  

    An approach to the two-microphone speech enhancement problem is discussed. Specifically, a maximum-likelihood (ML) problem is formulated for estimating the parameters needed for canceling the noise, and solved by the iterative EM (estimate-maximize) technique. The EM algorithm has been implemented for both a simplified and a more general scenario. The results improve upon those obtained with the classical least-squares approach View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Theory of order statistic filters and their relationship to linear FIR filters

    Page(s): 275 - 287
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1160 KB)  

    Necessary and/or sufficient conditions on both the filter coefficients and the signal process are derived in order that nonrecursive order statistic (OS) and linear filtering are equivalent operations. The results indicate that an understanding of OS filters hinges on a better understanding of the properties of signals containing logically monotonic components. The results extend a number of previous theories characterizing the well-known median and ranked-order filters to a broader class of filters and input signals View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimal deconvolution based on polynomial methods

    Page(s): 217 - 226
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (724 KB)  

    The problem of estimating the input to a known linear system is treated in a shift operator polynomial formulation. The mean-square estimation error is minimized. The input and a colored measurement noise are described by independent ARMA (autoregressive moving average) processes. The filter is calculated by performing a spectral factorization and solving a polynomial equation. The approach can be applied to input prediction, filtering, and smoothing problems as well as to the use of prefilters in the quadratic criterion. It applies to nonminimum-phase as well as unstable systems, as illustrated by two examples View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An analysis of several partially adaptive beamformer designs

    Page(s): 192 - 203
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (888 KB)  

    A general analysis of broadband interference cancellation in adaptive beamformers is provided. The results are then used to analyze beam, eigenstructure, and power-minimization approaches to the design of linearly constrained partially adaptive beamformers. It is shown that although beam-based designs are capable of good performance in narrowband environments, they are not appropriate for broadband environments. Eigenstructure-based designs are capable of excellent performance in narrow- or broadband environments. However, performance degrades if the number of adaptive weights is less than the eigenstructure dimension. Power-minimization designs are based on directly minimizing interference output power. Simulations indicate that the power-minimization design is capable of the best cancellation performance with a limited number of adaptive weights View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Input and output index mappings for a prime-factor-decomposed computation of discrete cosine transform

    Page(s): 237 - 244
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (584 KB)  

    A formal direct derivation of the prime-factor-decomposed computation algorithm is presented. The derivation is direct in the sense that it is based on the real cosine function without resort to the discrete Fourier transform expressions or the complex functions. Based on the equations obtained from the derivation, input and output index mappings are introduced in the form of tables. This tabulation enables any prime-factor-decomposable discrete cosine transform (DCT) to be implemented in a straight-forward manner. The use of the index mapping tables is demonstrated for the 12-point DCT View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Velocity filtering of acoustic well logging waveforms

    Page(s): 265 - 274
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1012 KB)  

    A velocity filter that is based on a particularly convenient and effective combination of an up/down filter and slanting (time-shift) operations is presented. The up/down filter is based on a cascade of spatial and temporal Hilbert transform filters. The Hilbert transform filters and the time-shift operators are finite-duration impulse response (FIR) filters designed according to the minimax criterion. The velocity filter is applied to acoustic well logging waveforms to isolate the initial P, S, and Stoneley waves from their reflected and converted waves at bed boundaries. To achieve this objective, the waveforms are time-shifted to align the primary wave of interest before the velocity filter is applied. Examples demonstrate the effectiveness of this procedure in separating the primary waves from their reflected and converted waves View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Efficient impulsive noise suppression via nonlinear recursive filtering

    Page(s): 303 - 306
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (328 KB)  

    A nonlinear recursive filter for the suppression of impulsive noises is proposed. The filter selects from each window the sample closest in value to the most recent output and is thus named the last output reference (LOR) filter. A relationship between the LOR and recursive median filters is derived, and some statistical properties are studied through computer simulation. The results indicate that this filter preserves edges while suppressing impulsive noise. It is shown that LOR filters are more effective in suppressing impulses and are often simpler to implement than median filters View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Quiver diagrams and signed adaptive filters

    Page(s): 227 - 236
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (808 KB)  

    A graphical technique for exploring the behavior of deterministic discrete time adaptive algorithms excited by periodic inputs is examined. The constraint to periodic inputs is convenient; the underlying concepts used also apply to stationary inputs drawn from finite alphabets. The resulting graph is based on plotting single-period parameter trajectories in the parameter error space and is essentially a discrete version of flow diagrams associated with continuous-time systems. For the least-mean-square (LMS) adaptive FIR filter with the algorithm step size sufficiently small, the change of adapted parameters from a particular initial value is given by the total update over the input period with the parameterization frozen at its initial value. This results in the familiar elliptical hyperparaboloid-surface steepest-descent interpretation. This method can be extended to the LMS adaptive infinite-input-response (IIR) filter with suitable approximations based on a small step size View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A robust spectral estimation by modeling an estimated autocovariance with an ARMA model

    Page(s): 181 - 191
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (620 KB)  

    The development, analysis, simulation results, and comparison of techniques for estimating spectral peaks of stationary signals in environments with low signal-to-noise ratio are discussed. Unlike traditional parametric methods, the proposed method estimates parameters of a model which best approximates the estimated autocovariance lag rather than the received signal. This technique is studied and evaluated in three different respects: using a performance index in the spectral domain, suppression of the moving-average (MA) portion and in terms of effective signal-to-noise ratio. This estimation technique demonstrates outstanding robustness and resolution for estimating both spectral peaks and amplitudes of multiple sinusoids embedded in white Gaussian noise, compared to traditional methods. When used in conjunction with Cadzow's autoregressive moving-average (ARMA) method using singular value decomposition (SVD), the technique extracts frequencies and amplitudes of existing sinusoids down to -17 dB while the ARMA method alone achieves only -10 dB on average View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new window and comparison to standard windows

    Page(s): 298 - 301
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (264 KB)  

    A new window, w(t)=0.62-0.48|t|+0.38 cos 2πt, |t|⩽1/2, is introduced. It is shown that the window is of a group with similar but different tradeoff properties with the Bartlett, the Hanning, and the Hamming windows. The proposed window is more effective than the Bartlett and the Hanning in the near sidelobes, and more effective than the Bartlett and the Hamming in the far sidelobes View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A dynamic programming algorithm for elastic registration of distorted pictures based on autoregressive model

    Page(s): 288 - 297
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (852 KB)  

    It is shown how to register a picture and a map when the deformation of the picture is severe by totally unknown. The method makes use of the combination of autoregressive (AR) modeling of the deformation working at the pixel level and a dynamic programming (DP) frame working at a higher level. AR modeling and DP work jointly: DP provides several sequences of observations which are candidates for an AR model, and at the same time the model furnishes a convenient cost function to DP. The only prerequisite to this method is that the map be expressible as an ordered sequence of primitives, and that a (not necessarily ordered) set of comparable primitives be extractable from the picture View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Detection of transient signals by the Gabor representation

    Page(s): 169 - 180
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (808 KB)  

    Gabor representation is used for the detection of transient signals with unknown arrival times. A one-sided exponential window function is used which seems to be most appropriate for transient modelling. Explicit expressions for the Gabor coefficients are given for this window function. When the given signal is random, so are the coefficients. The second-order moments of the Gabor coefficients are computed for a white noise signal. These are then used to introduce a detection statistic based on the Gabor coefficients. The proposed detector is capable of separating transients having different arrival times, even in this case where their waveforms partially overlap 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