Search

This paper presents a procedure for designing a robust controller which is able to stabilize a multivariable system with interval uncertainties in the system parameters. The uncertain parameters are assumed to lie within known intervals. The fundamental idea behind the procedure is to characterize the set of robust controllers as a convex set by the new matrix inequalities result. A suitable controller can then be chosen by means of an optimization algorithm. The controller thus chosen will be able to ensure that the closed-loop matrix lies within “desired” intervals for all the prescribed interval uncertainties in the system parameters.

The lowering of surface temperature below that of the water temperature underneath a floating sea-ice sheet often results in thermal cracking. The formation of these cracks is investigated theoretically by modeling the ice sheet as a floating elastic plate with thermal properties that vary with temperature and salinity. Results are obtained by extending previous work of Evans and Untersteiner and a typical crack spacing on the order of 200 m is found for perennial sea ice.

In this note the authors extend their earlier work on adaptive control for time-varying systems to the stochastic case. Simulation shows that this scheme is able to track time-varying parameters more effectively than some previous schemes.

In this paper we present a technique for designing robustly stable controllers for multivariable systems with poorly known parameters. We assume that the system matrices {A, B, C} are interval matrices, i.e. their elements live within prescribed intervals, and design a stablizing controller to ensure that the resulting closed-loop system remains stable when A, B and C take arbitrary values within their intervals.

In this paper the authors extend their earlier work on robust multivariable regulator design to the special cases where the system matrices suffer single row or single column variations. The algorithm presents a new and practical approach to robustness design when the uncertainties in system parameters are known to be restricted to single row or single column.

This paper considers the problem of estimating the state of a dynamic system from measurements obtained via a digital link with finite data rate R. The structures of the optimal coder and estimator for Markovian systems are derived. In particular, it is shown that the optimal coder for a Gauss-Markov system consists of a Kalman filter, followed by a stage which encodes the current Kalman estimate according to the symbols previously transmitted. A new suboptimal coder-estimator for linear systems is then constructed. Provided that a certain inequality linking the data rate to the dynamical parameters is satisfied, and under very mild assumptions on the noise distributions, this coder-estimator yields an expected absolute estimation error of the same order as in the classical situation with no data rate constraint. Hence if the classical estimation error approaches zero, then the rate-constrained error goes to zero at exactly the same speed.

This paper presents a novel, easily implementable set-theoretic solution to a multiple sensor data fusion and tracking problem. A centralized fusion architecture which performs data fusion at the measurement level is described. The algorithm combines 2D image data from two sensors to produce 3D tracks at each scan in the presence of false alarms. Track initiation and maintenance are performed as an integral part of the tracking process. >

This paper presents a new approach for multitarget tracking in a cluttered environment. Optimal all neighbour multi-target tracking (MTT) in clutter enumerates all possible joint measurement-to-target assignments and calculates the a posteriori probabilities of each of these joint assignments, e.g. J(I)PDA and MHT. The numerical complexity of this process is exponential in the number of tracks and the number of measurements involved. Our approach starts with an all-neighbour single-target tracking (STT) filter which provides the a priori probabilities of measurement origin; e.g. IPDA, IMM-PDA, ITS. These probabilities are used to modify the clutter density at the location of the selected measurements. In effect, the STT filter is transformed into a MTT filter with a numerical complexity which is linear in the number of tracks and the number of measurements. Measurement features, such as amplitude, can also be incorporated. Simulations are used to verify the approach when tracking crossing targets in an environment of heavy and non-uniform clutter.

We investigate the stability of an exact linearization style of control algorithm applied to a class of partially known nonlinear stochastic systems sampled at a constant rate. Since the system dynamics are not known, they are estimated using a filter which places very few assumptions on the underlying structure of the system. We show that when certain mild smoothness conditions are satisfied and the sampling rate is sufficiently fast, the control algorithm stabilizes the system in the sense that the closed-loop system becomes an ergodic Markov chain. Moreover, an explicit bound for the expected deviation of the system state from the origin is given.

A robust algorithm for estimating the parameters in a simple model of ionospheric electron density in the presence of model uncertainties and malicious noise is presented. Noisy sounding data is asynchronously recorded by a geographically distributed array of identical ionospheric sounders. The ionospheric dynamics are modelled as a state space system, with a robust H/spl infin/ type filter developed to track the evolving ionospheric state as new measurements become available.

We study a space-time precoding technique for MISO wireless systems by employing a linear prefilter at each transmit antenna. The channel is Rician fading, where the mean and variance of the propagation paths are known to the transmitter. This model includes the Rayleigh fading channels as special cases. We use channel capacity as the optimizing criterion for the prefilter design. This criterion provides a unified design of the prefilters for both Rician and Rayleigh fading channels. The optimum prefilters are functions of the channel mean and variance. The solution ranges from beamforming for Rician channels with high K-factor, to unitary diversity for Rayleigh fading cases, where delay diversity is an example. The MMSE equalizer is then used to detect the signal at the receiver. Analysis of bounds on error rate performance and numerical simulations for 4QAM input signals show significant diversity gains and array gains. The result also illustrates that having partial channel knowledge at the transmitter can strongly enhance the system performance.

This paper describes an off-line pulsewidth modulation (PWM) technique intended for the reduction of detrimental current harmonics in current source inverter systems. The procedure is based on the minimization of the amplitudes of offending harmonics through the determination of optimal-pulse positions using a gradient primal-dual algorithm.

A system can be termed non-minimum phase according to some definitions available in the literature and yet the same system may exhibit stable zero output constrained dynamics. We show that for non-affine nonlinear systems there may not exist a continuous control law which would keep the output identically equal to zero and for which the zero output constrained dynamics are stable, whereas a discontinuous controller which achieves this exists. We give conditions for existence and present a method for design of discontinuous switching controllers which yield stable zero dynamics. In this sense, the results of this paper enlarge the class of non-affine nonlinear systems that can be termed minimum-phase.

The use of a dead zone and a bounding ellipsoid for parameter estimation when measurement errors are bounded is discussed. the size of the dead zone is set to be exactly equal to the assumed noise bound. The algorithm retains the properties of computing parameter point estimates and allows a bounding ellipsoid to be computed at each iterative step.

We present a new method for on-line identification of time-varying FIR channels. Two conditionally coupled estimators are proposed. In both cases an augmented-state adaptive Kalman filter is employed for tracking the time-varying channel and estimating the mean channel response. Coupled to the Kalman filter is an algorithm for estimating the parameters of the underlying auto-regressive (AR) model which describes the time evolution of the channel. For the first coupled estimator, we propose a new recursive least squares algorithm for estimation of these AR parameters directly from the channel observations. An alternative algorithm based on estimation of the channel covariance is used in the second coupled estimator. A simulation example demonstrates the performance of the proposed estimators.

The method for current control of a photocathode source is described. This system allows for full remote control of a photocathode drive laser for resulting electron beam currents ranging from less than one microamp to a full current of five milliamps. All current modes are obtained by gating the drive laser with a series of electro-optical cells. The system remotely generates this control signal by assuming a mode of operation with the following properties selectable: Current mode as continuous or gated, micropulse density, macropulse gate width from single shot to 1 ms duration, macropulse synchronization to A/C line voltage (60 Hz) or an external trigger, 60 Hz phase and slewing through 60 Hz when applicable. All selections are derived from programmable logic devices operating from a master-oscillator resulting in a discrete, phase stable, pulse control for the drive laser. Complete system documentation is available at http://www.jlab.org/fel.

In this paper, we propose a new channel estimation algorithm for low Earth orbit satellite digital communications. Low Earth orbit satellite channels impart severe spreading in delay and Doppler on the transmitted signal. We model the channel as a tapped-delay-line filter with time-varying complex coefficients. The time evolution of the taps is described by an auto-regressive model. A coupled filter using recursive prediction error methods is proposed for estimating the parameters of the channel. Its performance is demonstrated via simulations.

In the past years, the problem of stabilising linear dynamical systems with low feedback data rates has been intensively investigated. A particular focus has been the characterisation of the infimum data rate for stabilisability, which specifies the smallest rate, in bits per unit time, at which information can circulate in a stable feedback loop. This paper extends this line of research to the case of fully-observed, finite-dimensional, linear systems without process noise but with control-independent, Markov parameters. Unlike previous formulations, the coding alphabet is permitted to be random and time-varying via a possible dependence on the observed Markov modes. Using quantisation techniques and real Jordan forms, it is shown that the smallest asymptotic mean data rate for stabilisability in r-th absolute output moment, over all coding and control schemes, is given by an exponent which measures the asymptotic mean growth rate of unstable eigenspace volumes. An explicit formula for it is obtained in the case of antistable dynamics. For scalar systems, this expression is quite different from an earlier one derived assuming a constant alphabet, in particular being independent of the moment order.

This paper presents new theorems on the theory of interval matrix inequalities and the theory of polynomials with interval roots, and applies them to the problem of robust pole-placement. We formulate optimization problems and derive convergent iterative algorithms which allow the designer to find controllers that place closed-loop poles within desired intervals for plants with unknown-but-bounded parameter uncertainties. The algorithms are computationally reasonable and provide a useful addition to currently existing control CAD tools.

This paper considers the problem of estimating the state of a remote system via a digital communication channel with finite capacity. Classical estimation theory does not apply since the estimator only observes the transmitted sequence of finite-valued symbols. Assuming no process or observation noise, a coding and estimation scheme is presented that generates an asymptotic mean-square error of zero, provided that the data rate and system parameters satisfy a certain condition. This scheme works for D-dimensional, time-varying, nonlinear systems that satisfy a Lipschitz-type condition, under very mild restrictions on the initial condition distribution. In the one-dimensional, linear time invariant case, the convergence condition reduces to an inequality that improves on previous results.

In this paper we measure and characterize the 55-65 GHz wireless channels for a typical desktop environment. Due to the presence of many obstructions and reflectors on the desktop the 60 GHz desktop channel differs from an intra-room propagation environment as described in the literature. The Saleh-Valenuela (S-V) model is used to model the desktop environment. Key S-V model parameters such as Cluster Decay Factor, Ray Decay Factor, Cluster Arrival Rate, and Ray Arrival Rate are extracted from the measured data

A new THz/IR/UV photon source at Jefferson Lab is the first of a new generation of light sources based on a Energy-Recovered, (superconducting) Linac (ERL). The machine has a 160 MeV electron beam and an average current of 10 mA in 75 MHz repetition rate hundred femtosecond bunches.These electron bunches pass through a magnetic chicane and therefore emit synchrotron radiation. For wavelengths longer than the electron bunch the electrons radiate coherently a broadband THz ∼ half cycle pulse whose average brightness is > 5 orders of magnitude higher than synchrotron IR sources. Previous measurements showed 20 W of average power extracted[1]. The new facility offers simultaneous synchrotron light from the visible through the FIR along with broadband THz production of 100 fs pulses with >200 W of average power (see G. P. Williams, this conference).The FELs also provide record-breaking laser power [2]: up to 10 kW of average power in the IR from 1 to 14 microns in 400 fs pulses at up to 74.85 MHz repetition rates and soon will produce similar pulses of 300–1000 nm light at up to 3 kW of average power from the UV FEL. These ultrashort pulses are ideal for maximizing the interaction with material surfaces. The optical beams are Gaussian with nearly perfect beam quality. See www.jlab.org/FEL for details of the operating characteristics; a wide variety of pulse train configurations are feasible from 10 microseconds long at high repetition rates to continuous operation.The THz and IR system has been commissioned. The UV system is to follow in 2005. The light is transported to user laboratories for basic and applied research. Additional lasers synchronized to the FEL are also available. Past activities have included production of carbon nanotubes, studies of vibrational relaxation of interstitial hydrogen in silicon, pulsed laser vapor deposition, nitriding of metals, and energy flow in proteins. This paper will present the status of the system and discuss some of the opportunities provided by this unique light source for modifying and studying materials.

This paper considers the identification of time-varying linear channels using maximum likelihood estimation. The channel is modelled as a tapped-delay line filter with complex coefficients. Due to the complexity of the likelihood function and the large number of parameters to be estimated, an analytical maximization of the likelihood function is infeasible. Therefore, gradient algorithms are considered. We consider the structure of the channel tap covariances, and present a gradient algorithm for finding the constrained maximum likelihood estimates of the channel parameters.

TJNAF recently commissioned its high-average-power infrared free-electron laser (FEL). It incorporates a superconducting accelerator that recovers about 75% of the electron-beam power and converts it to radio-frequency power. In achieving first lasing, the accelerator operated straight-ahead to deliver 38 MeV, 1.1 mA cw average current through the wiggler for lasing at wavelengths near 5 {micro}m. The waste beam was then sent directly to a dump. Stable operation at up to 311 W cw was achieved in this mode. Using a transport loop to send the waste electron beam back to the linac for energy recovery, the machine recently lased cw at up to 1720 W average power at 3.1 {micro}m, for which the electron-beam energy and average current were 48 MeV and 4.4 mA, respectively.

Optimum design for coherent optical OFDM transmitter has been analyzed. In contract to the direct-detection system, the optimal modulator bias point for the coherent system is pi where the nonlinearity and excessive loss are minimized.

We propose a simple and fast method for determining the direction-of-arrival of a mixture of narrowband and wideband sources. Our method uses the minimum eigenvector of the tempo-spatial correlation matrix as a weight vector in its spatial spectrum scanning pattern. It therefore does not involve determination of the effective rank of the tempospatial correlation matrix which is required by other existing high-resolution, direction-finding techniques.

The paper describes an algorithm developed for controlling access to a multiple-access broadcast channel. The algorithm, which is particularly simple, is suitable for networks with a small number of nodes. The algorithm was initially developed for use in a computer network dedicated to providing status information and control of a number of isolated plant sites. The channel available in this system was a radio-frequency troposcatter link, which supported a low 300 baud asynchronous data rate.