The application of self-organizing neural networks in processing nonlinear dynamic signals directly is investigated. The processing of a signal uses a model-based approach. The signal generating system is modeled by d...
详细信息
The application of self-organizing neural networks in processing nonlinear dynamic signals directly is investigated. The processing of a signal uses a model-based approach. The signal generating system is modeled by decomposing it into simpler subsystems and each subsystems is associated with a neuron on a single-layer network. Each subsystem is implemented using a temporally local linear combiner. The network is trained with a self-organizing procedure and the parameters of the linear combiners are updated by using the Widrow-Hoff adaptive rule. A competitive rule which takes into consideration the temporal dependence among the signal samples is presented. Simulation results are presented to illustrate the method.< >
Auditory evoked brainstem response (ABR) can be modeled as two closely spaced sinusoids immersed in noise. Conventional methods on the FFT are often not useful in the spectral estimation of short-duration ABRs. The au...
详细信息
Auditory evoked brainstem response (ABR) can be modeled as two closely spaced sinusoids immersed in noise. Conventional methods on the FFT are often not useful in the spectral estimation of short-duration ABRs. The authors propose the use of constrained adaptive notch filters (CANFs) implemented as a cascade of second-order sections, and discuss two special configurations based on CANF combined with a constrained notch filter (CNF) with fixed parameters. The problem associated with the convergence of the direct-form implementation of higher-order CANFs to a local minimum is also addressed. The methods proposed are shown to provide reliable spectral estimation and enhancement of the real data ABRs.< >
Least-squares (LS) estimation with a standard feedback neural network (SFBNN) which is based on an electrical model is investigated. In the energy function of a SFBNN, a non-quadratic term is included which is often n...
详细信息
Least-squares (LS) estimation with a standard feedback neural network (SFBNN) which is based on an electrical model is investigated. In the energy function of a SFBNN, a non-quadratic term is included which is often neglected while solving an optimization problem. It is shown that the non-quadratic term affects the solution of a continuous optimization problem. Properties of the non-quadratic term and the relation between the estimation error and several parameters of the SFBNN are discussed. A technique, called extended space iterative search (ESIS), is introduced to reduce the estimation error. Simulation results are presented to confirm the analysis result and the effectiveness of the proposed technique.< >
A distributed ray tracing technique which is based on the jittering concept of adding noise to the image is discussed. An octree data structure is used to improve the efficiency of calculation. By distributing the ray...
详细信息
A distributed ray tracing technique which is based on the jittering concept of adding noise to the image is discussed. An octree data structure is used to improve the efficiency of calculation. By distributing the rays in different dimensions-through the solid angle of light sources, over the lens, and in time-the effects of penumbras, depth of field, and motion blur are presented. The distributed ray tracing technique, is used to remove the aliasing error. Other effects, such as gloss and translucency, can also be produced by distributing the rays according to the specular distribution function and by distributing transmitted rays.< >
Statistical analysis of a fast algorithm for frequency estimation is presented. Details of how the analysis was derived with the aid of a symbolic algebra software package, Maple, are given, and some issues involved i...
详细信息
Statistical analysis of a fast algorithm for frequency estimation is presented. Details of how the analysis was derived with the aid of a symbolic algebra software package, Maple, are given, and some issues involved in the use of symbolic algebra packages in signalprocessing research are discussed.< >
The evolution graph monitor is defined and illustrated for on-line monitoring of computer-controlled processes with both discrete and continuous dynamics. Using a behavioral model specification of the process dynamics...
详细信息
The evolution graph monitor is defined and illustrated for on-line monitoring of computer-controlled processes with both discrete and continuous dynamics. Using a behavioral model specification of the process dynamics, the evolution graph monitor builds and maintain an efficient representation of the set of admissible system trajectories which are consistent with the on-line observations as they are received. In contrast to simulation-based methods, the evolution graph approach handles delayed, uncertain, and infrequent observations for systems with widely varying operating characteristics. A behavioral model specification and evolution graph monitor have been developed to perform on-line fault detection for an electric arc furnace in a continuous-caster steel mill. This application is used throughout the paper to illustrate the details of the approach.
This paper presents a method for detecting faults in systems with discrete observations. We examine systems representable by a behavioral model. The behavioral model framework characterizes the time evolution of syste...
详细信息
This paper presents a method for detecting faults in systems with discrete observations. We examine systems representable by a behavioral model. The behavioral model framework characterizes the time evolution of systems with discrete and continuous states and uncertainty in model parameters. Our method for detecting faults depends on an on-line encoding of the set of trajectories corresponding to a given observation and the dynamics specifications of the behavioral model. This encoding, called an evolution graph, is modified on-line as new observations are received. Faults are detected when the set of encoded trajectories is determined to be empty. This paper presents the modelling framework and discusses the construction and maintenance of the evolution graph.
To investigate the transition from the low density mode to the high density mode in an electron cyclotron resonance (ECR) discharge, a Langmuir probe and an E field probe were used to measure ion density and E field i...
To investigate the transition from the low density mode to the high density mode in an electron cyclotron resonance (ECR) discharge, a Langmuir probe and an E field probe were used to measure ion density and E field intensity as functions of axial position and power. The experiments were performed in argon at 0.13 Pa in a 7.9 cm diam cyclindrical source chamber propagating TE11 mode 2.45 GHz microwave power. Low mode was characterized by a standing wave throughout the plasma chamber and minimal power absorption. High mode exhibited nearly complete power absorption and no standing wave past the ECR zone. A sliding short (ss) was used to determine if the position of an E field null in the source chamber affected the transition between these two modes for various magnetic field configurations. The ss position had little effect on mode transition, relative power absorption or ion density when positioned downstream from a broad, large volume resonance zone (resonance near the mirror midplane). However, the plasma could not be ignited if the short was placed at or upstream from the large volume resonance *** the magnetic field was adjusted to yield a sharp, small volume resonance zone (resonance midway between the midplane and the throat), then positioning the ss to force an E field null at the resonance zone would prevent plasma ignition, even at 800 W forward power. The ion density exhibited a hysteresis (i.e., a direct jump from no plasma to a plasma density observed at other sliding short positions for that same forward power, thereafter following the ion density versus forward power dependence observed at nearby ss positions) on power cycling when the sliding short was within ∼1 cm of the broad resonance zone or within ∼1 cm of forcing a null in the sharp resonance zone caes. The ion density versus power curve for the broad volume case exhibited a change in slope over the transition region (Pforward ≊40 W) when the ion density at the resonance zone was ≊1×1011 cm
The debugging and testing of commercial software constitutes a substantial portion of the development time and, by most estimates, over fifty percent of the overall cost of software development. For critical applicati...
详细信息
作者:
A.K. OjhaD.B. KochCommunications
Information and Signal Processing (CISP) Group Department of Electrical and Computer Engineering University of Tennessee Knoxville TN USA
The authors investigate the performance of PN (pseudonoise) codes and complementary codes in an additive white Gaussian noise (AWGN) environment from a detection standpoint, and they provide a comparison of the perfor...
详细信息
The authors investigate the performance of PN (pseudonoise) codes and complementary codes in an additive white Gaussian noise (AWGN) environment from a detection standpoint, and they provide a comparison of the performance of these two coding methods in a pulsed radar context. A general method of choosing a cost function is proposed to compare the two coding schemes. Simulation results are presented which demonstrate that, for the AWGN channel model used, the complementary coded signals can provide improved performance over PN coded signals in practical pulsed radar systems.< >
暂无评论