A delay-dependent approach to robust H/sub /spl infin// filtering for uncertain discrete-time state-delayed systems

A delay-dependent approach to robust H/sub /spl infin// filtering is proposed for linear discrete-time uncertain systems with multiple delays in the state. The uncertain parameters are supposed to reside in a polytope and the attention is focused on the design of robust filters guaranteeing a prescribed H/sub /spl infin// noise attenuation level. The proposed filter design methodology incorporates some recently appeared results, such as Moon's new version of the upper bound for the inner product of two vectors and de Oliveira's idea of parameter-dependent stability, which greatly reduce the overdesign introduced in the derivation process. In addition to the full-order filtering problem, the challenging reduced-order case is also addressed by using different linearization procedures. Both full- and reduced-order filters can be obtained from the solution of convex optimization problems in terms of linear matrix inequalities, which can be solved via efficient interior-point algorithms. Numerical examples have been presented to illustrate the feasibility and advantages of the proposed methodologies.     

Genetic algorithm approach to fixed-order mixed H/sub 2//H/sub /spl infin// optimal deconvolution filter designs

The mixed H/sub 2//H/sub /spl infin// optimal deconvolution filter is proposed to achieve the H/sub 2/ optimal reconstruction and a desired robustness against the effect of uncertainties in signal processing from the H/sub /spl infin// norm perspective. However, the conventional mixed H/sub 2//H/sub /spl infin// optimal design filters are very complicated and are not practical for industrial applications. For simplicity of implementation and conservation of operation time, the fixed-order mixed H/sub 2//H/sub /spl infin// optimal deconvolution filter design is interesting for engineers in signal processing from the practical design perspective. In this study, to avoid the trap of local minima, a design method based on the genetic algorithm is introduced to treat the nonlinear optimization design problem of the fixed-order mixed H/sub 2//H/sub /spl infin// deconvolution filter. The convergence property of our design algorithm is also discussed. Finally, an example is presented to illustrate the design procedure and confirm the robustness performance of the proposed method.     

Induced l/sub 2/ and generalized H/sub 2/ filtering for systems with repeated scalar nonlinearities

This paper provides complete results on the filtering problem for a class of nonlinear systems described by a discrete-time state equation containing a repeated scalar nonlinearity as in recurrent neural networks. Both induced l/sub 2/ and generalized H/sub 2/ indexes are introduced to evaluate the filtering performance. For a given stable discrete-time systems with repeated scalar nonlinearities, our purpose is to design a stable full-order or reduced-order filter with the same repeated scalar nonlinearities such that the filtering error system is asymptotically stable and has a guaranteed induced l/sub 2/ or generalized H/sub 2/ performance. Sufficient conditions are obtained for the existence of admissible filters. Since these conditions involve matrix equalities, the cone complementarity linearization procedure is employed to cast the nonconvex feasibility problem into a sequential minimization problem subject to linear matrix inequalities, which can be readily solved by using standard numerical software. If these conditions are feasible, a desired filter can be easily constructed. These filtering results are further extended to discrete-time systems with both state delay and repeated scalar nonlinearities. The techniques used in this paper are very different from those used for previous controller synthesis problems, which enable us to circumvent the difficulty of dilating a positive diagonally dominant matrix. A numerical example is provided to show the applicability of the proposed theories.     

A polynomial approach to optimal and adaptive filtering withapplication to speech

Explicit polynomial solutions to the Wiener filtering problem are given. They rely on the identification of innovations models for the disturbance and for the noisy signal. The Wiener filter is found from the solution of a diophantine equation. Results illustrating the attenuation of background interference in a speech signal are presented. The explicit approach presented does not rely on minimizing a prediction error over a performance surface and can be applied where two input techniques are impracticable or impossible     

H/sub /spl infin// filtering for multiple-time-delay measurements

The aim of this paper is to study the H/sub /spl infin// filtering problem for linear continuous-time systems with both current and multiple-time-delay measurements. The key technique applied for deriving the filter is the reorganized innovation analysis approach in Krein space. A necessary and sufficient condition for the existence of an H/sub /spl infin// filter is derived, and the solution to the H/sub /spl infin// filter is given in terms of solutions of Riccati and matrix differential equations. An example is finally presented to show the effectiveness of the proposed approach.     

Robust H/sub /spl infin// filtering for Uncertain2-D continuous systems

This paper considers the problem of robust H/sub /spl infin// filtering for uncertain two-dimensional (2-D) continuous systems described by the Roesser state-space model. The parameter uncertainties are assumed to be norm-bounded in both the state and measurement equations. The purpose is the design of a 2-D continuous filter such that for all admissible uncertainties, the error system is asymptotically stable, and the H/sub /spl infin// norm of the transfer function, from the noise signal to the estimation error, is below a prespecified level. A sufficient condition for the existence of such filters is obtained in terms of a set of linear matrix inequalities (LMIs). When these LMIs are feasible, an explicit expression of a desired H/sub /spl infin// filter is given. Finally, a simulation example is provided to demonstrate the effectiveness of the proposed method.     

Robust H/sub 2/ filtering for a class of systems with stochastic nonlinearities

This paper addresses the robust H/sub 2/ filtering problem for a class of uncertain discrete-time nonlinear stochastic systems. The nonlinearities described by statistical means in this paper comprise some well-studied classes of nonlinearities in the literature. A technique is developed to tackle the matrix trace terms resulting from the nonlinearities, and the well-known S-procedure technique is adopted to cope with the uncertainties. A unified framework is established to solve the addressed robust H/sub 2/ filtering problem by using a linear matrix inequality approach. A numerical example is provided to illustrate the usefulness of the proposed method.     

Reduced-order H/sub /spl infin// filtering for stochastic systems

This paper deals with the reduced-order H/sub /spl infin// filtering problem for stochastic systems. Necessary and sufficient conditions are obtained for the existence of solutions to the continuous-time and discrete-time problems in terms of certain linear matrix inequalities (LMIs) and a coupling nonconvex rank constraint condition. Furthermore, when these conditions are feasible, an explicit parametrization of all desired reduced-order filters corresponding to a feasible solution is given. In particular, when the reduced-order filter is restricted to be a static one, then simple conditions expressed by LMIs only without any rank constraints are derived, and a parametrization of all solutions is also given. Finally, an illustrative example is provided to show the effectiveness of the proposed approach.     

Prediction of intrauterine pressure from electrohysterography using optimal linear filtering

We propose a method of predicting intrauterine pressure (IUP) from external electrohysterograms (EHG) using a causal FIR Wiener filter. IUP and 8-channel EHG data were collected simultaneously from 14 laboring patients at term, and prediction models were trained and tested using 10-min windows for each patient and channel. RMS prediction error varied between 5-14 mmHg across all patients. We performed a 4-way analysis of variance on the RMS error, which varied across patients, channels, time (test window) and model (train window). The patient-channel interaction was the most significant factor while channel alone was not significant, indicating that different channels produced significantly different RMS errors depending on the patient. The channel-time factor was significant due to single-channel bursty noise, while time was a significant factor due to multichannel bursty noise. The time-model interaction was not significant, supporting the assumption that the random process generating the IUP and EHG signals was stationary. The results demonstrate the capabilities of optimal linear filter in predicting IUP from external EHG and offer insight into the factors that affect prediction error of IUP from multichannel EHG recordings.     

Digital filtering using polynomial transforms

We define discrete transforms in a ring of polynomials. These polynomial transforms have the circular convolution property and can be used for the fast computation of 2-dimensional cyclic convolutions. This yields efficient algorithms for the implementation of 1- and 2-dimensional digital filters.     

a-Si:H TFTs using low-temperature CVD of Si/sub 3/H/sub 8/

a-Si:H layers have been deposited by chemical vapour deposition (CVD) at 350 degrees C using Si/sub 3/H/sub 8/ as the source gas. Inverted staggered gate thin-film transistors (TFTs) were fabricated with plasma-CVD-grown SiN/sub x/ as the gate insulator. Electron field-effect mobilities of 0.45 cm/sup 2//Vs were obtained and the on/off ratio in the drain current was 10/sup 6/.<>     

Fixed-order robust H/sub /spl infin// filter design for Markovian jump systems with uncertain switching probabilities

This paper discusses the fixed-order robust H/sub /spl infin// filtering problem for a class of Markovian jump linear systems with uncertain switching probabilities. The uncertainties under consideration are assumed to be norm-bounded in the system matrices and to be elementwise bounded in the mode transition rate matrix, respectively. First, a criterion based on linear matrix inequalities is provided for testing the H/sub /spl infin// filtering level of a filter over all the admissible uncertainties. Then, a sufficient condition for the existence of the fixed-order robust H/sub /spl infin// filters is established in terms of the solvability of a set of linear matrix inequalities with equality constraints. To determine the filter, a globally convergent algorithm involving convex optimization is suggested. Finally, a numerical example is used to illustrate that the developed theory is more effective than the existing results.     

Prefiltering approach for optimal polynomial prediction

A prefiltering approach for optimal prediction of polynomial signals is proposed. The new scheme enables the use of an arbitrary stable prefilter for which an optimal FIR postfilter is designed such that polynomial signals of given order are predicted unchanged. Additional degrees of freedom are used for noise suppression. The advantages of the approach are demonstrated with examples employing a first-order recursive prefilter     

Robust H/sub /spl infin// controller design with recurrent neural network for linear synchronous motor drive

In this paper, a robust controller design with H/sub /spl infin// performance using a recurrent neural network (RNN) is proposed for the position tracking control of a permanent-magnet linear synchronous motor. The proposed robust H/sub /spl infin// controller, which comprises a RNN and a compensating control, is developed to reduce the influence of parameter variations and external disturbance on system performance. The RNN is adopted to estimate the dynamics of the lumped plant uncertainty, and the compensating controller is used to eliminate the effect of the higher order terms in Taylor series expansion of the minimum approximation error. The tracking performance is ensured in face of parameter variations, external disturbance and RNN estimation error once a prespecified H/sub /spl infin// performance requirement is achieved. The synthesis of the RNN training rules and compensating control are based on the solution of a nonlinear H/sub /spl infin// control problem corresponding to the desired H/sub /spl infin// performance requirement, which is solved via a choice of quadratic storage function. The proposed control method is able to track both the periodic step and sinusoidal commands with improved performance in face of large parameter perturbations and external disturbance.     

On optimal linear filtering for edge detection

In this paper, we revisit the analytical expressions of the three Canny's (1983) criteria for edge detection quality: good detection, good localization, and low multiplicity of false detections. Our work differs from Canny's work on two essential points. Here, the criteria are given for discrete sampled signals, i.e., for the real, implemented filters. Instead of a single-step edge as input signal, we use pulses of various width. The proximity of other edges affects the quality of the detection process. This is taken into account in the new expressions of these criteria. We derive optimal filters for each of the criteria and for any combination of them. In particular, we define an original filter which maximizes detection and localization and a simple approximation of the optimal filter for the simultaneous maximization of the three criteria. The upper bounds of the criteria are computed which allow users to measure the absolute and relative performance of any filter (exponential, Deriche (1987), and first derivative of Gaussian filters are evaluated). Our criteria can also be used to compute the optimal value of the scale parameter of a given filter when the resolution of the detection is fixed.     

H/sub /spl infin// filtering for fuzzy singularly perturbed systems with pole placement constraints: an LMI approach

This paper addresses the problem of designing an H/sub /spl infin// filter for a class of nonlinear singularly perturbed systems described by a Takagi-Sugeno (TS) fuzzy model. Based on a linear matrix inequality (LMI) approach, we develop a fuzzy H/sub /spl infin// controller that guarantees that i) the L/sub 2/-gain from an exogenous input to a filter error is less than or equal to a prescribed value and ii) the poles of each local filter are within a prespecified region. In order to alleviate the ill-conditioning resulting from the interaction of slow and fast dynamic modes, solutions to the problem are given in terms of linear matrix inequalities, which are independent of the singular perturbation /spl epsiv/, when /spl epsiv/ is sufficiently small. The proposed approach does not involve the separation of states into slow and fast ones, and it can be applied not only to standard but also to nonstandard singularly perturbed nonlinear systems. A numerical example is provided to illustrate the design developed in this paper.     

MOS Characteristics of synthesized HfAlON-HfO/sub 2/ stack using AlN-HfO/sub 2/

We demonstrate a top-surface aluminized and nitrided HfO/sub 2/ gate dielectric using a synthesis of ultrathin aluminum nitride (AlN) and HfO/sub 2/. The reaction of AlN with HfO/sub 2/ through a subsequent high-temperature annealing incorporates Al and N into an HfO/sub 2/ layer, which results in a synthesis of HfAlON near the top surface of HfO/sub 2/, forming an HfAlON-HfO/sub 2/ stack structure. This approach suppresses interfacial layer growth and improves thermal stability of the dielectric, resulting in significant improvement in leakage current. It also shows no adverse effects caused by N and Al incorporation at the bottom interface.     

H/sub /spl infin// control of differential linear repetitive processes

Repetitive processes are a distinct class of two-dimensional (2-D) systems (i.e., information propagation in two independent directions) of both systems theoretic and applications interest. They cannot be controlled by direct extension of existing techniques from either standard [termed one-dimensional (1-D) here] or 2-D systems theory. Here, we give new results on the relatively open problem of the design of control laws using an H/sub /spl infin// setting. These results are for the sub-class of so-called differential linear repetitive processes which arise in applications.