By Topic

Acoustics, Speech and Signal Processing, IEEE Transactions on

Issue 5 • Date May 1990

Filter Results

Displaying Results 1 - 23 of 23
  • Matrix pencil method for estimating parameters of exponentially damped/undamped sinusoids in noise

    Page(s): 814 - 824
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (865 KB)  

    A study of a matrix pencil method for estimating frequencies and damping factors of exponentially damped and/or undamped sinusoids in noise is presented. Comparison of this method to a polynomial method (SVD-Prony method) shows that the matrix pencil method and the polynomial method are two special cases of a matrix prediction approach and that the pencil method is more efficient in computation and less restrictive about signal probes. It is found through perturbation analysis and simulation that, for signals with unknown damping factors, the pencil method is less sensitive to noise than the polynomial method. An expression of the Cramer-Rao bound for the exponential signals is presented.<> View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A class of iterative signal restoration algorithms

    Page(s): 778 - 786
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (728 KB)  

    A class of iterative signal restoration algorithms is derived based on a representation theorem for the generalized inverse of a matrix. These algorithms exhibit a first or higher order of convergence, and some of them consist of an online and an offline computational part. The conditions for convergence, the rate of convergence of these algorithms, and the computational load required to achieve the same restoration results are derived. An iterative algorithm is also presented which exhibits a higher rate of convergence than the standard quadratic algorithm with no extra computational load. These algorithms can be applied to the restoration of signals of any dimensionality. The presented approach unifies a large number of iterative restoration algorithms. Based on the convergence properties of these algorithms, combined algorithms are proposed that incorporate a priori knowledge about the solution in the form of constraints and converge faster than previously published algorithms View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • New polyphase filter bank-based analysis/synthesis systems

    Page(s): 876 - 880
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (368 KB)  

    Analysis/synthesis systems based on polyphase filter banks are presented for the 1-D and 2-D cases. Bandpass filters required in subband channels are constructed from nonsymmetric frequency translations of a prototype low-pass filter G(ω). Constraints on the frequency response G(ω) for perfect aliasing cancellation and signal reconstruction are then derived. An efficient structure consisting of a polyphase network and the fast Fourier transform (FFT) is proposed for realizing an analysis/synthesis system with an arbitrary number of subband channels and is shown to be more computationally efficient than other systems based on polyphase filter banks. The 1-D theory can be extended in two dimensions. The constraints on the frequency response G1, ω2) of a 2-D prototype low-pass filter are similar to those of the 1-D case if G1, ω2 ) is nonseparable. Computer simulations confirming the theoretical work are shown View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The unit step window and the identification of damped sinusoids

    Page(s): 764 - 768
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (428 KB)  

    Digital signals are found to have an `in-plane' representation of their z-transform zeros. This property allows for the location of the small and large zeros from short sections of signals, including very long signals or ones with severe zero clusters which have hitherto confounded all other methods of analysis. Also, it is shown how the location of the small zeros makes it possible to obtain highly improved identification of hidden heavily damped sinusoids View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A recursive algorithm for simultaneous identification of model order and parameters

    Page(s): 884 - 886
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (500 KB)  

    A recursive identification algorithm for SISO CARMA systems is presented based on an augmented information matrix (AIM). Decomposition of the AIM using UDU factorization provides simultaneous, recursive estimates of both the system parameters and the loss functions from order 0 to n, where n is the maximum possible order of the real process and U and D are upper and diagonal matrices, respectively. The most appropriate model order is then determined by examination of the loss functions. This approach results in a computationally efficient and numerically robust algorithm for systems of unknown or variable model order View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Weighted averaging of a set of noisy images for maximum signal-to-noise ratio

    Page(s): 890 - 895
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (764 KB)  

    The problem of estimating a signal from a weighted average of N registered noisy observations is considered. A set of optimal weighting coefficients is determined by maximizing a signal-to-noise ratio criterion. This solution can be computed by first standardizing each observation with respect to its first and second moments and then evaluating the first eigenvector of the corresponding N× N inner-product matrix. The resulting average is shown to be proportional to the first basis vector of the Karhunen-Loeve transform provided that the data has been standardized in an appropriate fashion. The low sensitivity of this approach to the presence of outliers is illustrated by using real electron micrographs of ostensibly identical virus particles View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Asymptotic statistics for a generalized frequency-wavenumber estimator

    Page(s): 804 - 813
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (656 KB)  

    The estimation of the frequency-wavenumber spectrum form the output of a sensor array placed in a homogeneous random field is considered. A generalized estimator which includes the Bartlett and minimum-variance estimators is used. Joint asymptotic normality of this frequency-wavenumber estimator is established; a precise asymptotic expression for the covariance matrix of the limiting distribution is obtained. It is shown that the generalized estimator converges in the mean-square sense. Confidence intervals based on the asymptotic normality results are presented for two array configuration examples View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Gradient-based adaptive IIR notch filtering for frequency estimation

    Page(s): 769 - 777
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (608 KB)  

    The use of gradient-based algorithms with infinite impulse response (IIR) notch filtering for estimating sinusoids imbedded in noise is investigated. Two notch filter model structures are presented. The first is for applications where two signal sources with correlated noise components can be assessed. The second can be used in situations where only one composite signal source is available. Error surface analysis indicates that second-order modules are unimodal and result in guaranteed convergence. Higher-order modules are multimodal and require judicious choice of initial parameter estimates. Simulation results are included to demonstrate the performance characteristics of both filter structures View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The prediction error of autoregressive small sample models

    Page(s): 858 - 860
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (228 KB)  

    In order selection, one single realization of a stochastic process is used twice, for the estimation of parameters for different model orders and for the selection of the best model order. The purpose of order selection is to find the model order that gives the best fit to other realizations of the same stochastic process. This fit is expressed by the squared prediction error and it will increase if too many parameters are used. The weak parameter criterion (WPC) is an estimate for the squared prediction error, with the special feature that it is computed from the same observations that are used for the estimation of the parameters. A novel justification for the principle of the WPC is presented, which shows the correspondence between the WPC and the squared prediction error. Calibration formulas are presented that describe the averages over many simulation runs of WPC, the squared prediction error and residual variance all as a function of the order of the estimation model View full abstract»

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

    Page(s): 825 - 838
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (948 KB)  

    AN adaptive algorithm is developed for online estimation of the poles of autoregressive (AR) processes. The method estimates the poles directly from the data without intermediate estimation of the AR coefficients or polynomial factorization. It converges rapidly, is computationally efficient, and attains the Cramer-Rao bound (CRB) asymptotically. A closed-form expression for the asymptotic CRB is provided. Convergence to the true solution is proved, and methods are discussed for extending the algorithm for use with more general (e.g. autoregressive moving-average) models. Numerical examples are presented to demonstrate the performance of the algorithm View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Real-time speech segmentation using pitch and convexity jump models: application to variable rate speech coding

    Page(s): 741 - 748
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (568 KB)  

    Convexity jump models are combined with period identification in the time domain to provide an efficient segmentation technique adapted to the needs of coders. The algorithm presented works online, with no training, and is speaker-independent. Structural problems found in previous systems (blank detection spots, long decision delays) have been overcome. Both the simplicity and the generality (information theory basis for the divergence test) allow applications of the automatic online segmentation method in other areas of signal processing View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The effect of spatial averaging on spatial correlation matrices in the presence of coherent signals

    Page(s): 880 - 884
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (472 KB)  

    Two types of spatial averaging are considered: subaperture averaging and redundancy averaging. The reduction of coherence due to subaperture averaging is shown to depend on the separation between the sources: highly coherent, closely spaced sources remain highly coherent after subaperture averaging. In finite averaging situations, this remaining high degree of coherence will prevent adaptive direction-finding methods such as MUSIC or EV from resolving the two signals. In contrast, redundancy averaging greatly reduces intersignal coherence regardless of source spacing, but bias may be introduced into the source bearing estimates View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Efficient iterative methods for FIR least squares identification

    Page(s): 887 - 890
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (312 KB)  

    Two iterative methods based on matrix splitting are presented to facilitate the least-squares identification of finite impulse response (FIR) systems. The first method yields an order-recursive solution to a previously proposed fixed-order iterative procedure. The second is a simplified and computationally efficient method and is applicable when the input signals are white, or nonwhite with a correlation coefficient of value less than 1/3. Convergence performance of both algorithms is briefly discussed. Simulation results indicate satisfactory performance of both methods View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • AR model identification with unknown process order

    Page(s): 872 - 876
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (400 KB)  

    A method for simultaneous autoregressive (AR) model order selection and identification is proposed, which is based on the adaptive Lainiotis filter (ALF). The method is not restricted to the Gaussian case, is applicable to online/adaptive operation, and is computationally efficient. It can be realized in a parallel processing fashion. The AR model order selection and identification problem is reformulated so that it can be fitted into the framework of a state space under uncertainty estimation problem framework. The ALF is briefly presented and its application to the specific problem is discussed. Simulation examples are presented to demonstrate the superior performance of the method in comparison with previously reported ones View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Serial architectures for the implementation of 2-D digital filters and for template matching in digital images

    Page(s): 853 - 857
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (320 KB)  

    Serial architectures are derived for the hardware realization of two-dimensional (2-D) digital filters (IIR and FIR), and for the implementation of template matching in digital images. The architectures utilizes two's-complement representation and table look-up techniques. The contents of the lookup tables is generated via hardware from the filter coefficients. The architectures require a modest amount of hardware, and could be suitable for VLSI implementation View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Orthogonal sets of data windows constructed from trigonometric polynomials

    Page(s): 870 - 872
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (256 KB)  

    Suboptimal, easily computable substitutes for the discrete prolate spheroidal windows used by D.J. Thomson (Proc. IEEE, vol.70, p.1055-1096, 1982) for spectral estimation are given. Trigonometric coefficients and energy leakages of the window polynomials are tabulated View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimal hydrophone placements under random perturbations

    Page(s): 860 - 864
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (388 KB)  

    To improve the performance of an array of hydrophones used to locate an acoustic source, an approach is proposed for hydrophone placement which minimizes a conventional bearing or range variance bound averaged over the random deviation of sensor positions. The Cramer-Rao lower bounds are used on localization error covariances to distribute the sensors in an effective manner for line and towed arrays. The source localization problem for M sensors is formulated. A lower bound on the estimate of the localization parameters from the measurements is examined. This is then applied to a simple line array and to a towed array for various levels of sensor position uncertainty. The results are utilized to distribute the sensors in approximately optimal fashion View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Design issues and an architecture for the monolithic implementation of a parallel digital signal processor

    Page(s): 839 - 852
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1356 KB)  

    Design issues in the implementation of a parallel digital signal processor for measurement systems are discussed. Constraints imposed by the application and VLSI technology are shown to lead to the design criteria of modularity, flexibility, programmability, and high speed. An architecture for the monolithic implementation of a parallel digital signal processor based upon these criteria is then presented. The use of asynchronous, self-timed function units each consisting of an instruction queue, data queues, and combinational logic is supported. Data and control interlocks are used for synchronization within the processor to facilitate both fine and coarse-grain parallelism. A programming example and simulation results are presented for a finite impulse response (FIR) filter. These results confirm the digital signal processor's efficiency. Using a conventional synchronous design, the simulation of a single prototype processor is presented. This design requires six clock cycles to compute a radix-2 fast Fourier transform (FFT) butterfly compared to 12 instruction cycles for a TI TMS320C30. Compared to a Motorola DSP 56001, the prototype processor permits a faster instruction cycle and less I/O overhead. It is estimated that a 1024-point FFT can be sped up 10 times using 10 processors View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Analysis of a frequency-domain adaptive IIR filter

    Page(s): 864 - 870
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (416 KB)  

    Convergence of frequency-domain adaptive pole-zero IIR (infinite impulse response) filter is studied. The algorithm is shown to converge in probability to an associated ordinary differential equation (ODE) which in turn converges to a local minimum of its performance surface. An analysis of the performance surface shows that the algorithm converges to one of N-factorial members in an equivalence class of global minimum points, where N is the number of adaptive poles. Saddle points exist on manifolds that separate members in the equivalence class. This explains `shoulders' in the MSE convergence curves and also suggests one way of avoiding these shoulders which cause slow convergence. A second-order simulation example confirms the above results View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The use of data dependence graphs in the design of bit-level systolic arrays

    Page(s): 787 - 793
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (572 KB)  

    The use of bit-level systolic array circuits as building blocks in the construction of larger word-level systolic systems is investigated. It is shown that the overall structure and detailed timing of such systems may be derived quite simply using the dependence graph and cut-set procedure developed by S.Y. Kung (1988). This provides an attractive and intuitive approach to the bit-level design of many VLSI signal processing components. The technique can be applied to ripple-through and partly pipelined circuits as well as fully systolic designs. It therefore provides a means of examining the relative tradeoff between levels of pipelining, chip area, power consumption, and throughput rate within a given VLSI design View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An adaptive offset cancellation technique for adaptive filters

    Page(s): 799 - 803
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (332 KB)  

    Offset may be introduced in the adaptation process in certain implementations of adaptive filters. The adaptive filters. The adaptive filter still converges with the offset, however, to nonoptimal tap weights with excess mean square error. Here, an offset cancellation technique, also adaptive, is proposed. A set of new taps is introduced to counteract the offset. The convergence of this adaptive cancellation technique is demonstrated analytically and by using computer simulation View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Detection and estimation for multiple targets with two omnidirectional sensors in the presence of false measurements

    Page(s): 749 - 763
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (836 KB)  

    A track-before-detect methodology for target detection and estimation in the presence of false measurements is presented that uses two omnidirectional passive sensors. The estimation technique is based on maximum-likelihood estimation. The measurement model is nonlinear and includes false alarms. The algorithm is first developed for a single target and then extended to multiple targets. For multiple targets, unresolved measurements are also considered to provide a realistic analysis of targets crossing in the measurement space. The Cramer-Rao lower bound is derived for the target parameter estimation in the presence of false measurement. A detection mechanism that can validate the existence of a target corresponding to the estimated track is formulated. For a single target, it is shown that only the global maximum leads to the acceptance of the target hypothesis. The test for multiple targets is obtained by formulating a multiple-hypotheses problem. The theoretical performance predictions are validated via Monte Carlo simulations. The effect on the performance of the density of false measurements is illustrated in examples. The highest false-measurement density for which this technique works corresponds to SNR=2 dB View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Frequency translation using variable delay

    Page(s): 794 - 798
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (396 KB)  

    It is shown how, using clocked delay systems, it is possible to translate the frequency of an applied periodic voltage. The translation can represent an increase or a decrease in frequency. Moreover, the amount of the translation can itself be made to change with time. A clocked delay system is constructed in a straight-forward manner using shift registers. The theory is tested using clock rates that change linearly with time and using clock rates that provide a linear time variation of delay. The resulting frequency translation is measured for sinusoidal, triangular and square wave inputs. The results are in good agreement with the theory 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