Showing papers on "Kernel adaptive filter published in 1979"

An Optimal Frequency Domain Filter for Edge Detection in Digital Pictures

Abstract: Edge detection and enhancement are widely used in image processing applications. In this paper we consider the problem of optimizing spatial frequency domain filters for detecting edges in digital pictures. The filter is optimum in that it produces maximum energy within a resolution interval of specified width in the vicinity of the edge. We show that, in the continuous case, the filter transfer function is specified in terms of the prolate spheroidal wave function. In the discrete case, the filter transfer function is specified in terms of the sampled values of the first-order prolate spheroidal wave function or in terms of the sampled values of an asymptotic approximation of the wave function. Both versions can be implemented via the fast Fourier transform (FFT). We show that the optimum filter is very effective for detecting blufred and noisy edges. Finally, we compare the performance of the optimum edge detection filter with other edge detection filters using a variety of input images.

A convergence proof for a hyperstable adaptive recursive filter (Corresp.)

Abstract: Hyperstability, a concept from nonlinear stability theory, is used to develop a real-time adaptive recursive filter useful in a nonstationary environment.

Analysis of the frequency domain adaptive filter

Abstract: The purpose of this note is to demonstrate significant analytical simplifications for studying the behavior of adaptive filtering in the frequency domain as opposed to studying the behavior of adaptive filtering in the time domain. A closed form expression, for the single complex weight in the frequency domain adaptive filter, is presented which allows significant statistical analysis to be performed. The mean-square error of the filter is evaluated as a function of the algorithm step size and the signal and noise powers.

Statistical design of autoregressive-moving average digital filters

Abstract: Procedures are presented for the systematic design of digital filters that contain poles and zeros. The procedures are simple, fast, and effective. All of the important algorithms are of the Levinson-type. The first key idea in the paper is that one may begin a design by posing a linear prediction problem for a stochastic sequence. The second is that a high-order "whitening" filter may be constructed for this sequence and "inverted" to yield a high-order all-pole filter whose spectrum approximates the spectrum of the stochastic sequence. The third key idea is that the all-pole filter may be used to generate consistent unit-pulse and covariance sequences for use in the Mullis-Roberts algorithm. This algorithm is then used to obtain a low-order digital filter, with poles and zeros, that approximates the high-order all-pole filter. The results demonstrate that the Mullis-Roberts algorithm, together with the design philosophy of this paper, may be used with profit to reduce filter or stochastic model complexity and to design spectrum-matching digital filters.

Adaptive Clutter Filtering Using Autogressive Spectral Estimation

Abstract: A new adaptive filter to reject clutter is derived using autoregressive spectral analysis techniques. The adaptive filter performs open. Ioop processing, resulting in a shorter transient response, and is therefore suitable for radar waveforms containing only a small number of samples. A number of examples including application to ballistic missile defense are presented to demonstrate the performance capabilities of the new adaptive filter.

Self-adaptive Kalman filter

Abstract: It is shown how the refined instrumental variable (r.i.v.) method of recursive parameter estimation can be modified simply so that it functions as an optimal adaptive filter and state-estimation algorithm.

A sequential failure detection approach and the identification of failure parameters

Abstract: This paper is concerned with the problem of a failure diagnosis for a discrete-time system with parametric failure, in which the occurrence time and mode of parametric failure cannot be estimated in advance. The failure diagnosis system which is proposed consists of three parts : (i) a normal mode filter, (ii) a detector for anomaly states, and (iii) an adaptive extended Kalman filter. The normal mode filter is called the optimal Kalman filter and transports the information of its innovation sequence to the detector. The detector which is based on the SPRT approach detects anomaly states affected by the parametric failure. The adaptive extended Kalman filter estimates simultaneously system parameters and the states under the failure mode. The adaptive procedure is directed by increasing the calculated covariance on the basis of hypothesis tests for the estimation errors of unknown parameters. Numerical results for a simple plant model illustrate the effectiveness of the proposed failure diagnosis system.

High Order Algorithms for Adaptive Filters

Abstract: Faster convergence of adaptive filters has been of particular interest in the areas of adaptive equalizers and adaptive antennas. The algorithm most frequently suggested is some variation of the firstorder gradient-descent LMS algorithm. This paper investigates a general procedure for the design of higher-order algorithms. Convergence of the mean weight vector and the variance is compared for three typical algorithms. Results indicate that high order convergence of the mean weight vector can easily be achieved, and this in itself can be useful. However, the variance response of the high order algorithms can tend to have an offsetting effect which may preclude their use in some applications.

A constrained minimum power adaptive beamformer with time-varying adaptation rate

Abstract: A linear adaptive algorithm was developed for array beamforming purposes. The design goal for the algorithm is to minimize the squared filter output subject to filter constraints which allow energy propagating from the array steering direction to pass without being distorted. The adaptive filter coefficients were initialized to satisfy the constraints which were preserved during the iterations. The adaptation rate is inversely varied with filter output and total input channel power. Performance of the algorithm was studied using the recorded short‐period array data from the Korean Seismic Research Station. Processed were a high‐amplitude signal from Kamchatka, a medium‐amplitude signal from eastern Kazakh, and a number of low‐amplitude signals from central Eurasia. Results of signal‐to‐noise ratio gain relative to a conventional beamformer among the events tested were consistent and were in the range of 4.5 to 6.5 dB in the wide passband. Much better signal‐to‐noise ratio improvement was obtained in the l...

A Stochastic Filter for Transverse Section Reconstruction

Abstract: A new algorithm, based on minimum-mean-square reconstruction error, involves modification of the filter kernel based on the signal-to-noise of each projection target. This algorithm gives improved resolvability relative to other algorithms for the resolution phantom chosen for simulations with Poisson noise.

The ideal averaging filter: its applications and realizations

Abstract: The optimum filter for improving the signal/noise ratio in a commonly-encountered instrumentation situation is the ideal averaging filter. Various ways of realizing the ideal response are discussed and rational approximants of order 1 to 6 are given. The circuits required to implement these approximants are considerably less complex than those required for the alternative realizations.

A short-term sequential regression algorithm

Abstract: A short-term sequential regression (STSER) formulation is introduced to facilitate adaptive filtering in nonstationary environments. A corresponding STSER algorithm for finite impulse response (FIR) filters is derived. Experimental results involving an STSER predictor are presented. For the purpose of comparison, corresponding results using other available adaptive algorithms are also included.

A Sampled Analog MOS LSI Adaptive Filter

Abstract: A novel realization of an adaptive filter using sampled analog MOS LSI techniques is described in this paper. The basic functional block of the adaptive filter is an electrically programmable transversal filter whose tap weights are modified according to the least mean square algorithm. A comparison has been made among different available approaches for implementing a programmable transversal filter. A rotating tap weight structure [1] has been used to realize a 32-tap programmable transversal filter with features for the adaptive operation included on an NMOS silicon-gate chip. The adaptive filter has been characterized as a parameter estimator which finds application in echo cancellation and noise cancellation. A wide range of magnitude and phase characteristics of the unknown system (to be modeled by the adaptive filter) have been used to test the performance of the present adaptive system. Results on the residual error and the convergence time under different conditions are reported. Some practical limitations of the system are also presented.

Adaptive noise cancelling of a sinusoidal interference using a lattice structure

Abstract: Lattice structures have recently been proposed for application to such areas as adaptive linear prediction, adaptive noise cancelling, and adaptive equalization. In this paper, we will consider the suppression of a sinusoidal interference by the use of an adaptive lattice structure implemented as a noise canceller. The performance of the adaptive lattice will be analyzed and compared with that of an LMS transversal adaptive noise canceller. Experimental results will also be presented and discussed.

Optimal estimation of ship's attitudes and attitude rates

Abstract: Estimation of ship rotational motions induced by ocean waves plays a central role in many navigation and fire control applications. Inertial-type instruments are available which give good measurements of the attitude, but some form of signal processing is necessary to obtain angular rates or predict attitudes. Using optimal moving-average (or transversal) filters, we can obtain very good estimates of attitude rates as well as predictions of these values. Filter parameters can be changed adaptively to maintain good performance as the ship changes heading or velocity. The problem of designing these optimal filters is examined in detail and numerical results are given for a particular set of conditions. Two implementations of the adaptive filter are discussed. One is based on a recursive estimation of the process autocorrelations with the filter coefficients being recomputed at periodic intervals or whenever nonstationary conditions are detected. The second implementation is based on Widrow's LMS algorithm. Both alternatives for the adaptive filter implementation are quite reasonable in terms of their computational requirements. The steady-state performance analysis can be considered to be a lower bound on the errors incurred by an adaptive filter.

Charge-Coupled Device (CCD) Adaptive Discrete Analog Signal Processing

Abstract: A CCD adaptive signal processor is described which uses a so-called "clipped-data" least mean square (LMS) error algorithm to optimize the selection of tap weights in a CCD filter. A detailed description of a 16-tap monolithic silicon CCD analog adaptive filter is also presented. The filter is comprised of a basic linear combiner formed with a nondestructively tapped CCD analog delay line and electrically reprogrammable MOS analog conductances as the tap weights. Two methods of varying the analog conductance are discussed: 1) variable V GS with fixed threshold voltage V T and 2) variable V T with fixed V GS . The former is performed with a CCD bidirectional charge control weight adjustment, whereas the latter is accomplished with MNOS memory transistors. To demonstrate the feasibility of adaptive analog signal processing, a 2-tap weight CCD adaptive filter is described and experimental results presented. Applications include optimum filtering, prediction, noise cancellation, and system modeling.

2-D FIR filter design from independent "Small" generating kernels using a mean square and Tchebyshev error criterion

Abstract: We present and compare two design techniques for a large class of 2-D FIR filters. They can be implemented by means of sequential convolutions with small size kernels. Design methods are based upon the minimization of either the mean square or the maximum error between the synthesized filter and a design prototype in the Fourier domain.

Linear Filtering on the Surface of a Sphere.

Abstract: : Although methods of filtering have been developed for representing on a planar surface the sub-grid scale process of diffusion in numerical modeling of the atmosphere, the proper form of filtering for a spherical domain remains to be selected. In this study, the characteristics of several different forms of a highly scale dependent low pass filter are examined and compared for the case where the filter is applied to a scalar field on the surface of a sphere. The phase and amplitude response functions of the various forms of the filter indicate that the simplest form, although it does not preserve area-weighted mean values, approaches most closely the criteria established for the ideal filter. This indication is verified by the test computations in which each form of the filter is applied up to 10,000 times to a noisy scalar field. (Author)

An Adaptive Distributed-Measurement Extended Kalman Filter for a Short Range Tracker. Volume II.

Abstract: : An adaptive Extended Kalman Filter algorithm is designed to track a distributed (elliptical) source target in a closed loop tracking problem, using outputs from a forward looking infrared (FLIR) sensor as measurements. The filter adaptively estimates image intensity, target size and shape, dynamic driving noise, and translational position changes due to two effects: actual target motion, and atmospheric jitter. Atmospheric backgrounds are studied for the effect of temporal and spatial correlations on filter performance. A Monte Carlo analysis is conducted to determine filter performance for two target scenarios: approximately straight approach and cross range constant velocity. Good performance is obtained for the first two trajectories. For the second trajectory, a one sigma tracking error of .2 pixel (4 microrad) with a signal to noise ratio of 12.5. The filter adapts well to changes in image intensity, size, and shape. (Author)

Parameter estimation and digital filter techniques

Abstract: The dynamic control of fermentation by the use of predictive models requires data upon substrate and reagent flows, and their uptake rates, which can be used in material balance calculations.’,* However, available plant data contain measurement errors that can seriously affect such calculations unless corrected. A number of papers in recent years have dealt with the adjustment of experimental data to achieve material balance.’-s However, these methods have been applied to historical rather than current data. Other authors have described the use of extended Kalman filtering techniques for parameter estimation and batch process control!” Although these authors worked only with simulated systems and used large-scale computers, the Kalman filtering method used is a recursive technique capable of operating on each new data point and being programmed into a small process control computer. Unfortunately, it has been shown that Kalman filter estimation of process states is seriously degraded by inaccuracies in the process Thus its use in control may be limited until more accurate models of fermentation processes are developed. The technique can be used with excellent results for the estimation of environmental variables, where the processes are physical, rather than biologic, and the models are well known. The estimation of weights, temperatures, gas concentrations, substrate flow rates, etc. can be significantly improved by the use of this method. In this respect digital filter technology can contribute to the formulation of accurate models and can thus pave the way for the more general application of such models to fermentation process control.

Noise cancellation and inverse filtering using a compact high-performance c.c.d. adaptive filter

Abstract: This letter describes results obtained from a compact adaptive filter based on a charge-coupled device, programmable, analogue transversal filter. The adaptive algorithm used was the Widrow l.m.s. algorithm and results are presented for the 64-filter-point system used as a self-tuning filter, a noise canceller and as an inverse filter. The results illustrate the high quality of performance possible using this structure.

FIR filter hardware reduction with adaptive delta-modulation

Abstract: This paper presents a new type of digital system which makes use of adaptive delta-modulation for input digitization and is particularly suited to the hardware implementation of small, low power, programmable digital filters. An experimental prototype was built which demonstrated a finite impulse response (FIR) digital filter. This filter has no requirement for hardware multipliers and has a significant reduction in input sample storage capacity when compared to equivalent pulse code modulation (PCM) filters. Quantitative as well as qualitative results show that this new filter is especially useful with voice inputs.

Detection of sinusoids by linear prediction filter frequency response

Abstract: Four processors for the detection of a sinusoid of known frequency but unknown phase in Gaussian white noise of known spectrum level are described: the likelihood ratio detector; a processor employing incoherently averaged discrete Fourier transforms (DFTs); and two systems utilizing the frequency response function of an adaptive linear prediction filter. Power curves for these receivers are compared, showing that the adaptive detectors are similar in performance to the incoherent DFT processor. An expression for the optimal adaptive filter feedback constant is derived.

Adjustable model reference adaptive control

Abstract: An adjustable model reference adaptive control strategy is proposed where both the controller and reference model parameters are adapted. Adaptive control laws for this structure implementing pole cancellation and replacement and avoiding the zero cancellation usually required by model reference adaptive control are developed from self-tuning adaptive algorithm modification.

Hybrid realisation of an adaptive filter for real-time noise-cancelling applications

Abstract: An efficient hybrid realisation of an adaptive filter for real-time noise-cancelling applications is described. The filter is based on an analogue tapped delay line but its coefficients are digital and are updated according to a simplified l.m.s. algorithm. A single multiplying d.a. device is used for performing the multiplications. The structure of the filter, its performance and processed examples are presented.

Nonrecursive digital filter designs by means of Korovkin kernels

Abstract: In this paper, the general class of Korovkin kernels is introduced as window functions in the design of FIR digital filters. In particular, the utilization of the Jackson kernel in window designs is discussed in detail. The Jackson window is applied to the design of a low-pass filter with a transition region. The advantages and limitations of the Jackson window designs are discussed.

An adaptive linear function elimination filter

Abstract: This paper presents an analysis of a linear function elimination filter ( KEF) with adaptive feedback decision applied to the coherent detection ASK-system. Using the adaptive feedback decision permits a more precise estimation of the predictor coefficient, whose value depends on the input signal spectrum. The theoretical analysis of adaptive feedback decision system is based on the finite state Markov chain. An estimation of the probability of error is given in comparison with the probability of error for such a filter without a feedback decision system. The experimental results of testing this filter for detecting the binary ASK signal in the presence of both sinusoidal interference and gaussian noise confirm the obtained theoretical results.

Performance Analysis of Adaptive Processors via Diffusion Model Approximations and the Fokker-Planck Equation

Abstract: Some results on adaptive array processors and adaptive filters that have been obtainsd by the use of diffusion model approximations and the Fokker-Planck equation are described.

Spline filter for optimum phase demodulation

Abstract: An optimum nonlinear filter is realized by sequentially updating the spline coefficients of the relevant conditional distribution. The nonlinear filtering problem considered is that of phase demodulation with a two-dimensional phase process model. Speed and accuracy comparison of spline realization with other realizations, such as Fourier filter and point mass, will be provided

CCD adaptive filtering for robust LPC speech processing

Abstract: A monolithic CCD adaptive filter chip is described which implements the Widrow-Hoff "clipped-data" LMS adaptive algorithm. The chip can be used as a pre-filter noise canceller, analysis filter, or pre-whitener for a pitch extractor in linear prediction coding (LPC) voice bandwidth reduction systems.