Radiation and scattering from ferrite-tuned cavity-backed slotantennas: theory and experiment

A three-dimensional finite-element method hybridized with the spectral/spatial domain method of moments is presented for the analysis of ferrite-tuned cavity-backed slot antennas. The cavity, which is partially filled with magnetized ferrite layers, is flush mounted on an infinite ground plane with possible dielectric or magnetic overlay. The antenna operates primarily in the ultrahigh-frequency band. The finite-element method is used to solve for the electric-field distribution inside the cavity, whereas the spectral-domain approach is used to solve for the exterior region. An asymptotic extraction of the exponential behavior of the Green's function followed by a spatial evaluation of the resulting integral is used to improve computational speed. Radar cross section, input impedance, return loss, gain, and efficiency of ferrite-tuned cavity-backed slots (CBS) are calculated for various biasing conditions. Numerical results are compared with experimental data     

Learning and convergence analysis of neural-type structurednetworks

A class of feedforward neural networks, structured networks, has recently been introduced as a method for solving matrix algebra problems in an inherently parallel formulation. A convergence analysis for the training of structured networks is presented. Since the learning techniques used in structured networks are also employed in the training of neural networks, the issue of convergence is discussed not only from a numerical algebra perspective but also as a means of deriving insight into connectionist learning. Bounds on the learning rate are developed under which exponential convergence of the weights to their correct values is proved for a class of matrix algebra problems that includes linear equation solving, matrix inversion, and Lyapunov equation solving. For a special class of problems, the orthogonalized back-propagation algorithm, an optimal recursive update law for minimizing a least-squares cost functional, is introduced. It guarantees exact convergence in one epoch. Several learning issues are investigated.     

Dynamic adhesive force measurements under vertical and horizontal motions of interacting rough surfaces

An instrument to measure dynamic adhesive forces between interacting rough surfaces has been developed. It consists of four parts, namely, main instrument body, vertical positioning system with both micrometer and nanometer positioning accuracies, horizontal positioning system with nanometer positioning accuracy, and custom-built high-resolution, and high dynamic bandwidth capacitive force transducer. The vertical piezoelectric actuator (PZT) controls the vertical (approaching and retracting) motion of the upper specimen, while the horizontal PZT controls the horizontal (reciprocal) motion of the lower specimen. The force transducer is placed in line with the upper specimen and vertical PZT, and directly measures the adhesive forces with a root-mean-square load resolution of 1.7 microN and a dynamic bandwidth of 1.7 kHz. The newly developed instrument enables reliable measurements of near-contact and contact adhesive forces for microscale devices under different dynamic conditions. Using the developed instrument, dynamic pull-in and pull-off force measurements were performed between an aluminum-titanium-carbide sphere and a 10 nm thick carbon film disk sample. Three different levels of contact force were investigated; where for each contact force level the vertical velocity of the upper sample was varied from 0.074 to 5.922 microms, while the lower sample was stationary. It was found that slower approaching and retracting velocities result in higher pull-in and pull-off forces. The noncontact attractive force was also measured during horizontal movement of the lower sample, and it was found that the periodic movements of the lower disk sample also affect the noncontact surface interactions.     

Passive Attack Detection for a Class of Stealthy Intermittent Integrity Attacks

This paper proposes a passive methodology for detecting a class of stealthy intermittent integrity attacks in cyber-physical systems subject to process disturbances and measurement noise. A stealthy intermittent integrity attack strategy is first proposed by modifying a zero-dynamics attack model. The stealthiness of the generated attacks is rigorously investigated under the condition that the adversary does not know precisely the system state values. In order to help detect such attacks, a backward-in-time detection residual is proposed based on an equivalent quantity of the system state change, due to the attack, at a time prior to the attack occurrence time. A key characteristic of this residual is that its magnitude increases every time a new attack occurs. To estimate this unknown residual, an optimal fixed-point smoother is proposed by minimizing a piece-wise linear quadratic cost function with a set of specifically designed weighting matrices. The smoother design guarantees robustness with respect to process disturbances and measurement noise, and is also able to maintain sensitivity as time progresses to intermittent integrity attack by resetting the covariance matrix based on the weighting matrices. The adaptive threshold is designed based on the estimated backward-in-time residual, and the attack detectability analysis is rigorously investigated to characterize quantitatively the class of attacks that can be detected by the proposed methodology. Finally, a simulation example is used to demonstrate the effectiveness of the developed methodology.      

Fault accommodation of a class of multivariable nonlinear dynamicalsystems using a learning approach

Presents a learning approach for accommodating faults occurring in a class of nonlinear multi-input-multi-output (MIMO) dynamical systems. Changes in the system dynamics due to a fault are modeled as unknown nonlinear functions of the measurable state variables. The closed-loop stability of the robust fault accommodation scheme is established using Lyapunov redesign methods. A simulation example, based on a model of a jet engine compression system, is used to illustrate the fault accommodation design procedure     

An analytical framework for local feedforward networks

Interference in neural networks occurs when learning in one area of the input space causes unlearning in another area. Networks that are less susceptible to interference are referred to as spatially local networks. To obtain a better understanding of these properties, a theoretical framework, consisting of a measure of interference and a measure of network localization, is developed. These measures incorporate not only the network weights and architecture but also the learning algorithm. Using this framework to analyze sigmoidal, multilayer perceptron (MLP) networks that employ the backpropagation learning algorithm on the quadratic cost function, we address a familiar misconception that single-hidden-layer sigmoidal networks are inherently nonlocal by demonstrating that given a sufficiently large number of adjustable weights, single-hidden-layer sigmoidal MLPs exist that are arbitrarily local and retain the ability to approximate any continuous function on a compact domain.     

Indirect adaptive nonlinear control of drug delivery systems

This paper investigates the use of adaptive neural network techniques for modeling and automatic control of mean arterial pressure through the intravenous infusion of sodium nitroprusside. An indirect model reference-based adaptive nonlinear control scheme with neural networks approximating the unknown nonlinearities, is developed. In this formulation nonlinear estimators are used to adaptively approximate the system uncertainty and augment the linear control law for improved performance. The overall design is based on self-tuning the controller to the specific response characteristics of individual patients. Computer simulations illustrate the ability of radial basis function networks to model the unknown nonlinearities and improve the closed-loop system characteristics     

Distributed Fault Diagnosis With Overlapping Decompositions: An Adaptive Approximation Approach

This technical note deals with the problem of designing a distributed fault detection methodology for distributed (and possibly large-scale) nonlinear dynamical systems that are modelled as the interconnection of several subsystems. The subsystems are allowed to overlap, thus sharing some state components. For each subsystem, a local fault detector is designed, based on the measured local state of the subsystem as well as the transmitted variables of neighboring states that define the subsystem interconnections. The local detection decision is made on the basis of the knowledge of the local subsystem dynamic model and of an adaptive approximation of the interconnection with neighboring subsystems. The use of a specially-designed consensus-based estimator is proposed in order to improve the detectability of faults affecting variables shared among different subsystems. Simulation results provide an evidence of the effectiveness of the proposed distributed fault detection scheme.