Showing papers on "Kernel adaptive filter published in 1983"

A unified approach to time- and frequency-domain realization of FIR adaptive digital filters

Abstract: Specific implementations of the finite impulse response (FIR) block adaptive filter in the frequency domain are presented and some of their important properties are discussed. The time-domain block adaptive filter implemented in the frequency domain is shown to be equivalent to the frequency-domain adaptive filter (derived in the frequency domain), provided data sectioning is done properly. All of the known time- and frequency-domain adaptive filters [1]-[12], [16]-[18] are contained in the set of possible block adaptive filter structures. Thus, the block adaptive filter is generic and its formulation unifies the current theory of time- and frequency-domain FIR adaptive filter structures. A detailed analysis of overlap-save and overlap-add implementations shows that the former is to be preferred for adaptive applications because it requires less computation, a fact that is not true for fixed coefficient filters.

Mel Log Spectrum Approximation (MLSA) filter for speech synthesis

Abstract: The spectral envelope of speech can be represented efficiently by the log magnitude spectrum on the nonlinear frequency scale, which is close to mel scale (called mel-log spectrum). the mel cepstrum defined by its Fourier coefficients is also considered to have a suitable property as the parameter to represent the spectral envelope. So far, no satisfactory filter has been reported for the synthesis approximating the mel-log spectrum. This paper presents a method of constructing the mel-log spectrum approximation (MLSA) filter, which has a relatively simple structure and a low coefficient sensitivity, together with a design example of MLSA filter for speech synthesis. the transfer function of MLSA filter is represented by Pade approximation, which approximates the exponential of the transfer function of the filter (so-called basic filter). Since the transfer function of the basic filter is represented by a polynomial with the transfer function of the first-order all-pass filter as the variable, it is necessary in the realization of the filter to delete from the feedback loop the path without a delay. By the construction method of MLSA filter shown in this paper, the path without delay can easily be deleted from the feedback loop in the MLSA filter. the obtained MLSA filter is of relatively simple structure and has low coefficient sensitivity. the quantization characteristics of the coefficient are also satisfactory.

On the effects of varying filter bank parameters on isolated word recognition

Abstract: The vast majority of commercially available isolated word recognizers use a filter bank analysis as the front end processing for recognition. It is not well understood how the parameters of different filter banks (e.g., number of filters, types of filters, filter spacing, etc.) affect recognizer performance. In this paper we present results of performance evaluation of several types of filter bank analyzers in a speaker trained isolated word recognition test using dialed-up telephone line recordings. We have studied both DFT (discrete Fourier transform) and direct form implementations of the filter banks. We have also considered uniform and nonuniform filter spacings. The results indicate that the best performance (highest word accuracy) is obtained by both a 15-channel uniform filter bank and a 13-channel nonuniform filter bank (with channels spacing along a critical band scale). The performance of a 7-channel critical band filter bank is almost as good as that of the two best filter banks. In comparison to a conventional linear predictive coding (LPC) word recognizer, the performance of the best filter bank recognizers was, on average, several percent worse than that of an eighth-order LPC-based recognizer. A discussion as to why some filter banks performed better than others, and why the LPC-based system did the best, is given in this paper.

Two-sided linear filter identification

Abstract: It is argued that there are a number of situations in life science where it is desirable to attempt a two-sided linear filter identification. A simple method is presented for the determination of a nonparametric, two-sided linear filter from system input and output data. The time-domain filter is determined from a matrix equation involving the input autocorrelation function and the two-sided cross-correlation function. The resulting filter minimises the sum of squared differences between the actual and predicted outputs.

Convergence behavior of the constant modulus algorithm

Abstract: An adaptive filter algorithm has been developed and introduced [1] for use with constant envelope waveforms, e.g., FM communication signals. It has proven capable of suppressing additive interferers as well as equalization, without the need for a priori statistical information. In this paper, aspects of dynamic convergence behavior are discussed, with conclusions supported by simulation.

Rejection of Pulsed CW Interference in PN Spread-Spectrum Systems Using Complex Adaptive Filters

Abstract: In this paper, the performance of direct sequence QPSK spread-spectrum systems using complex adaptive filters in the presence of pulsed CW interference is analyzed. Both adaptive prediction error filters and adaptive transversal filters with two-sided taps are considered. It is shown that the time constant of the tap weight adaptation in the interference off-interval is usually much greater than the time constant in the on-interval, and that this is beneficial for the system since it results in retaining the rejection property of the filter. Under steady-state conditions, the tap weights are calculated. Analytical expressions for the signal-to-noise ratio improvement under the least favorable interference condition are given. It is shown that the performance of the two-sided transversal filter is better than that of the prediction error filter.

Adaptive tracking of multiple hot-spot target IR images

Abstract: In the recent past, the capability of tracking dynamic targets from forward-looking infrared (FLIR) measurements has been improved substantially, by replacing standard correlation trackers with adaptive extended Kalman filters. This research investigates a tracker able to handle "multiple hot-spot" targets, in which digital (or optical) signal processing is employed on the FLIR data to identify the underlying target shape. This identified shape is then used in the measurement model portion of the filter as it estimates target offset from the center of the field-of-view. In this algorithm, an extended Kalman filter processes the raw intensity measurements from the FLIR to produce target estimates. An alternative algorithm uses a linear Kalman filter to process the position indications of an enhanced correlator in order to generate tracking estimates; the enhancement is accomplished not only by thresholding to eliminate poor correlation information, but also by incorporating the dynamics information from the Kalman filter and the on-line identification of the target shape as a template instead of merely using previous frames of data. The performance capabilities of these two algorithms are evaluated under various tracking environment conditions and for a range of choices of design parameters.

The Asymptotic Optimal Frequency Domain Filter for Edge Detection

Abstract: In an earlier paper by Shanmugam, Dickey, and Green, an edge detection filter was derived which maximized the energy within a specified interval about an edge feature. The initial expression of this filter involved a prolate spheroidal wave function. However, a careful analysis of the application of an asymptotic approximation to this function uncovered a major dimensional error. The corrected derivation of the asymptotic optimal filter forms the subject of this paper. To verify the results, the filter found is compared to a similar filter developed independently by Marr and Hildreth.

A fast convergence frequency domain adaptive filter

Abstract: This correspondence presents a new fast convergence algorithm for frequency domain adaptive filter and its applicability to acoustic noise cancellation in speech signals.

A Kalman Filter Approach to Precision GPS Geodesy

Abstract: The high stability of the GPS signals makes it possible to determine differential position over short baselines with an accuracy of the order of centimeters. This has been demonstrated using very long baseline interferometry (VLBI) methods of radio astronomy. This paper presents an alternative approach using Kalman filter methods. It is based on an adaptive filter concept developed by D. T. Magi11 in 1965. The scheme employs parallel Kalman filters with each filter being modeled for a different integer wavelength assumption. As the phase measurement sequence progresses, the adaptive scheme “learns” which Kalman filter corresponds to the correct hypothesis, and thus it both resolves the wavelength ambiguity and estimates differential position simultaneously. Simulated results are presented which indicate convergence in about 10 minutes of satellite observation time.

The effects of selected signal processing techniques on the performance of a filter-bank-based isolated word recognizer

Abstract: To implement an isolated word recognizer based on filter bank techniques, decisions must be made as to how to condition the speech signal prior to the filter bank analysis (preprocessing), how to condition the feature vector at the output of the filter bank analysis (postprocessing), and how to perform the time alignment in the pattern comparison between an unknown test pattern and previously stored reference patterns (registration and distance computation). In the past most designers of such word recognition systems made arbitrary choices about how the various signal processing operations were to be carried out. This paper presents results of a systematic study of the effects of selected signal processing techniques on the performance of a filter bank isolated word recognizer using telephone-quality speech. In particular, the filter bank analyzer was a 13-channel, critical-band-spaced filter bank with excellent time resolution (impulse response durations of from 3 to 30 ms) and poor frequency selectivity (highly overlapping filters with ratios of center frequency to 3-dB bandwidth of about 8 for each band). Among the signal processing techniques studied were: preemphasis of the speech signal; time and frequency smoothing of the filter bank outputs; thresholding, quantization, and normalization of the feature vector; principal components analysis of the feature vector; local and global distance computations for use in the time alignment procedure; and noise analysis in both training and testing. Each of the signal processing techniques was studied individually; hence no tests were run in which several of the techniques were used together. Results showed that some fairly simple signal processing operations provided the best overall performance in the noise-free case; in noisy conditions performance degraded significantly for signal-to-noise ratios less than about 24 dB.

An adaptive nonlinear digital filter with lattice orthogonalization

Abstract: A novel approach to nonlinear filtering with minimum mean square error criterion is presented. This method considers the class of nonlinear filters with Volterra series structures under the assumption that filter inputs are Gaussian, and a relatively simple solution results which is directly applicable in many practical situations. Moreover, two simple parameter adaption algorithms for the second order Volterra filter are presented and it is shown that their convergence speeds depend on the squared ratio of maximum to minimum eigenvalues of the input autocovariance matrix. Finally, the lattice orthogonalization of filter input is considered for faster convergence.

A new look at the Magill adaptive filter as a practical means of multiple hypothesis testing

Abstract: The Magill adaptive filter can be used to detect known signals in the presence of Gauss-Markov noise. In this application, the various hypotheses are accounted for outside the bank of Kalman filters, and thus all filters have the same gains and error covariances. This commonality makes it feasible to use the Magill scheme in large-scale multiple-hypothesis testing applications.

An adaptive approach for time delay estimation of band-limited signals

Abstract: This correspondence concerns the use of the Widrow's adaptive least-mean-square (LMS) adaptive filter algorithm [1] for estimating the time delay between two-sensor outputs. Theoretical bias and estimation error are discussed. This analytical study provides a clue for determining the most effective way of processing the received signals through the adaptive filter. Computer simulation results are also included.

Adaptive filter with a time-domain implementation using correlation cancellation loops.

Abstract: The results of a feasibility study of an optical adaptive filter are presented. The processor is a time-domain implementation using correlation cancellation loops. Included is a theoretical verification of the correlation cancellation loop approach for linear prediction. The processor architecture and performance are described in detail. The results are encouraging although limited by laboratory equipment.

Analysis and performance evaluation of an adaptive notch filter

Abstract: An adaptive notch filter is derived using a general prediction error framework. The proposed infinite impulse response filter has a special structure that guarantees the desired transfer characteristics. The filter coefficients are updated by a version of the recursive maximum likelihood algorithm.

Spatial lattice filter for high-resolution spectral analysis of array data

Abstract: In the analysis of time/frequency domain signals, lattice filters have proved popular, providing significant advantages over conventional tapped delay line digital filters. Improvements in adaptation times and model order identification have been observed, and the modular structure of these filters allows a straightforward VLSI hardware implementation. In the paper, a lattice-structured spatial filter for use in adaptive arrays is described. This filter is shown to take on a triangular structure in the spatial domain which provides a step-by-step solution to the prediction problem. Several uses of the lattice filter in array processing are then discussed, particularly the extraction of high-resolution spatial spectral estimates from the parameters of the filter. These spatial spectral estimators include the maximum-entropy, maximum-likelihood and eigenvector (Pisarenko) based methods and are used to obtain high-resolution maps of the directional power incident upon an array of spatially distributed sensors. It is also shown how multiple look direction constraints may be applied to the lattice filter, providing the capability for enhancing desired signals in the presence of directional interference.

Recursive self-tuning algorithm for adaptive Kalman filtering

Abstract: A new recursive algorithm for adaptive Kalman filtering is proposed The signal state-space model and its noise statistics are assumed to depend on an unknown parameter taking values in a subset [', '] of Rs The parameter is estimated recursively using the gradient of the innovation sequence of the Kalman filter The unknown parameter is replaced by its current estimate in the Kalman-filtering algorithm The asymptotic properties of the adaptive Kalman filter are discussed

A monolithic adaptive filter

Abstract: Reports a monolithic adaptive filter which has been realized using purely analog sampled-data MOS and CCD techniques. The filter implements a full Widrow least mean-squares algorithm over 65 data points. Central to this design is a novel, compact analog multiplier/accumulator circuit, which is presented in detail. The 65-point adaptive filter, which is cascadable, dissipates 200 mW from a 15 V supply, and operates at sample rates up to 100 kHz.

Convergence of partitioned adaptive filters for systems with unknown biases

Abstract: A convergence theorem on performance bounds for partitioned adaptive estimation is extended to include biases among the unknown parameters. The importance and implications of this rarely considered case are briefly discussed.

Tracking characteristics of the Kalman filter in a nonstationary environment for adaptive filter applications

Abstract: In this paper, the Kalman filter theory is used to develop an algorithm for updating the tap-weight vector of an adaptive tapped-delay line filter that operates in a nonstationary environment. The tracking behaviour of the algorithm is discussed in detail. Computer simulation experiments show that this algorithm, unlike the exponentially weighted recursive least-squares (deterministic) algorithm, is always stable. Simulation results are included in the paper to illustrate this phenomenon.

A new design for digital adaptive filters

Abstract: The design and operation of a digital adaptive filter based on a modified Widrow-Hoff algorithm is described. It uses LSI components as main elements. This filter realization converges faster than other realizations reported hitherto because its weight vector is updated at every sampling instant.

Rapid interference suppression using a Kalman filter technique

Abstract: Directional interference suppression is based on a recursive estimation procedure of the state of a filter forming the directional interference spectrum. For the stationary filter, this leads to an iterative solution of the Wiener-Hopf matrix equation. It can be shown that the number of samples required for convergence is related to the number of degrees of freedom of the filter. The nonstationary case of a changing interference environment is treated by an appropriate deterministic dynamic process model. Tracking performance and problems of digital implementation are studied by computer simulation.

Space-time duality in digital filter structures

Abstract: An FIR filter with fixed-point coefficients is viewed as a two-dimensional binary pattern of ones and zeros, as seen by the digital machine. The operator z-1represents shift in one dimension, time, and the operator 2-1represents shift in the other dimension, space. This two-dimensional binary approach results in two digital filter structures, the split in space and merge in time structure and its dual the split in time and merge in space structure. The two structures are optimized to minimize a particular computational complexity measure. Extensions and potential applications of this work are discussed.

State variable estimation using adaptive Kalman filter with robust smoothing

Abstract: The development of a conventional Kalman filter is based on full knowledge of system parameters, noise statistics and deterministic forcing functions. This work addresses the problem of known system parameters and unknown noise statistics and deterministic forcing functions. Two concepts are investigated: 1) adaptive weight functions for the Kalman filter gain and error covariance matrices, where these weights are functions of sample means and variances of the innovations sequence; and 2) robust smoothing of the estimated state variables. The concepts presented relative to this particular problem address the limited class of linear system dynamics with associated linear measurements. Nonlinear system dynamics with associated linear or nonlinear measurements, however, are not precluded. The concepts apply to those cases where the observations made by a sensor are the variables to be estimated. An application to a simple linear system is presented; however, primary application would be to the estimation of position, velocity and acceleration for a maneuvering body in three dimensional space based on observed data collected by a remote sensor tracking the maneuvering body. Estimates of the state variables using the adaptive process for the simple linear system during the periods when the system is not being forced are relatively close to those of the conventional Kalman filter for congruent periods, but there is some increase in mean square error because the adaptive estimator is no longer optimal. During periods when the system is being forced a vast improvement, as compared with those estimates of the conventional Kalman filter, is realized with the adaptive gain, covariance weight, and associated robust smoothing procedure. The estimates derived with the adaptive procedure during the periods of system forcing do, however, contain a considerable level of mean-square error. This seems to be a prevailing shortfall of adaptive estimation procedures. The tradeoff is knowledge of the deterministic forcing functions versus high mean-square estimate error in the absence of that knowledge.

An Optically Implemented Multiple-Stage Kalman Filter Algorithm

Abstract: A working paper is presented on the mathematical development and analysis of an optically implemented multiple-stage Kalman filter algorithm which uses two previously developed estimation models (linear first-order Gauss-Markov and constant turn-rate) for high energy laser pointing and tracking. An overview of the estimation models reveals model equivalence in mid- to long-range tracking applications and superiority of the constant turn-rate model (at the expense of a much higher computational burden) for both short-range and evasive target tracking. Real world constraints are to be forcibly imposed on the optical filter by limiting the choice of all system components to off-the-shelf units whose performance criteria are well characterized. Derivation of the filter architecture subject to the real world constraints shows the pielined iterative systolic array architecture to be significantly superior. Filter development based on this architecture is expected to generate a MTF which yields superior performance of the optical filter over its electronic counterpart based both on the output statistics produced and system throughput capability Additional analyses of filter performance reveal potential filter enhancement with the incorporation of range and relative velocity data obtained through use of a laser doppler velocimeter and an optical heterodyne detector. Current and planned future research efforts are also presented.© (1983) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.