The major purpose of this paper is to promote interchange between the fields of pattern recognition and communications, in the realm of statistical classification. The general class of second-order measures of quality...
详细信息
The major purpose of this paper is to promote interchange between the fields of pattern recognition and communications, in the realm of statistical classification. The general class of second-order measures of quality for statistical classification is defined. The variety of members in this class that have been used by practitioners or proposed by theorists for numerical pattern-classification and signal waveform-classification are compared and contrasted. The several measures that are the most generally applicable are shown to be either equivalent to each other or characterizable in terms of each other, thereby revealing an inherent unity. For example, the ratio of between-class-scatter to within-class-scatter used in pattern recognition and the ratio of signal-energy to noise-energy used in communications are unified through an identification of signal with between-class-scatter and noise with within-class-scatter. Results on equivalences are stated and proved for waveform classification rather than numerical classification in order to complement the extensive literature on the latter, and to emphasize applicability to communications. This entails introduction of a scatter ratio for waveforms. In a companion paper, second-order measures of quality are used as a basis for a general nearestprototype signal-classification methodology; canonical signal features for this methodology are identified, and a general approach for determining appropriate class prototypes is given. These two papers provide an integrated approach to the design of a complete signal classifier, i.e., feature extraction and discriminant-functional design tailored to fit a minimumdistance discrimination rule.
A comprehensive analysis of the mean-square learning characteristics of stochastic-descent algorithms is presented. The approach is based on the commonly exploited simplifying assumption of stationary independent trai...
A comprehensive analysis of the mean-square learning characteristics of stochastic-descent algorithms is presented. The approach is based on the commonly exploited simplifying assumption of stationary independent training vectors. Characteristics analyzed include stability, steady-state misadjustment, initial rate of convergence, optimum step size, and steady-state autocovariance and spectral characteristics of the weight-vector. Effects on these characteristics due to degree of randomness of stochastic gradient, particular data distribution, and data corruption are isolated and analyzed. An objective of the work is to keep the number of simplifying assumptions and approximations to a minimum. Comparison of results with previous more approximate analyses are made. Lernkurven von stochastischen Gradienten Algorithmen werden untersucht. Die vereinfachende Annahme von stationären, unabhängigen Trainingsvektoren wird benutzt. Charakteristiken die untersucht werden beinhalten Stabilität, ‘steady state’ Fehlanpassung, Start Konvergenz, optimaler Schritt sowie ‘steady state’ Autokovarianz und spektrale Charakteristik des Gewichtsvektors. Die Effekten auf diese Charakteristiken von dem Zufälligkeitsgrad des stichastischen Gradienten, besonderer Daten Verteilung und Daten Verderbung werden isoliert und analysiert. Ein Zielpunkt dieser Arbeit ist es die Anzahl vereinfachenden Annahmen und Approximationen auf ein minimum zu beschränken. Vergleiche mit Resultaten die von approximativeren Analysen stammen werden gezogen. Une analyse d'ensemble des caractéristiques de convergence de l'erreur quadratique moyenne dans les algorithmes utilisant la décroissance du gradient stochastique est presentée. Cette approche est basée sur l'hypothése simplificatrice classique de stationnarité et indépendance des vecteurs de test. Les caractéristiques analysées comprennent la stabilité, l'écart d'état stable, la vitesse initiale de convergence, le pas optimum d'incrémentation, et l'autocovarian
The implementation of digital filtering algorithms using pipelined vector processors is investigated. Modeling of vector processors and vectorization methods are explained, and then the performances of several impleme...
详细信息
The implementation of digital filtering algorithms using pipelined vector processors is investigated. Modeling of vector processors and vectorization methods are explained, and then the performances of several implementation methods are evaluated based on the model. Vector processor implementation of FIR filtering algorithms using the outer product method and the indirect convolution method is evaluated. Recursive and adaptive filtering algorithms, which lead to dependency problems in direct vector processor implementations, are implemented very efficiently using a newly developed vectorization method. The proposed method computes multiple output samples at a time, making the vector length independent of the filter order. Illustrative examples comparing theoretical results with Cray X-MP simulation results are included.
In this paper the authors derive simple approximate formulas for the performance of entropy-encoded DPCM for a Gaussian random process and a frequency-weighted mean-square distortion measure. Using these results they ...
详细信息
In this paper the authors derive simple approximate formulas for the performance of entropy-encoded DPCM for a Gaussian random process and a frequency-weighted mean-square distortion measure. Using these results they compare the performance of DPCM to the information theoretic rate-distortion bound. They study the effect on the performance of DPCM of the spectrum of the input process, the frequency weight in the distortion measure, and the number of prediction coefficients. They also examine briefly the case of achromatic still images using line-by-line and two-dimensional DPCM encoding with intrafield and intraframe information.
We consider a class of supervisory control problems that require infinite state supervisors Petri nets with inhibitor arcs (PN's) to model the supervisors. We compare this PN-based approach to supervisory control ...
详细信息
We consider a class of supervisory control problems that require infinite state supervisors Petri nets with inhibitor arcs (PN's) to model the supervisors. We compare this PN-based approach to supervisory control to automata-based approaches. The primary advantage of a PN-based supervisory controller is that a PN-based controller provides a finite representation of an infinite state supervisor. For verification, implementation, and testing reasons, a finite PN-based representation of an infinite state supervisor is preferred over an automata-based supervisor. We show that this modeling advantage is accompanied by a decision disadvantage, in that in general the controllability of a language that can be generated by the closed-loop system is undecidable.
A novel framework for parallel subgraph isomorphism on GPUs is proposed, named GPUSI, which consists of GPU region exploration and GPU subgraph matching. The GPUSI iteratively enumerates subgraph instances and solves ...
详细信息
A novel framework for parallel subgraph isomorphism on GPUs is proposed, named GPUSI, which consists of GPU region exploration and GPU subgraph matching. The GPUSI iteratively enumerates subgraph instances and solves the subgraph isomorphism in a divide-and-conquer fashion. The framework completely relies on the graph traversal, and avoids the explicit join operation. Moreover, in order to improve its performance, a task-queue based method and the virtual-CSR graph structure are used to balance the workload among warps, and warp-centric programming model is used to balance the workload among threads in a warp. The prototype of GPUSI is implemented, and comprehensive experiments of various graph isomorphism operations are carried on diverse large graphs. The experiments clearly demonstrate that GPUSI has good scalability and can achieve speed-up of 1.4–2.6 compared to the state-of-the-art solutions.
This paper presents a texture segmentation algorithm based on a hierarchical wavelet decomposition. Using Daubechies' four-tap filter, an original image is decomposed into three detail images and one approximate i...
详细信息
This paper presents a texture segmentation algorithm based on a hierarchical wavelet decomposition. Using Daubechies' four-tap filter, an original image is decomposed into three detail images and one approximate image. The decomposition can be recursively applied to the approximate image to generate a lower resolution of the pyramid. The segmentation starts at the lowest resolution using the K-means clustering scheme and textural features obtained from various sub-bands. The result of segmentation is propagated through the pyramid to a higher resolution with continuously improving the segmentation. The lower resolution levels help to build the contour of the segmented texture, while higher levels refine the process, and correct possible errors.
We analyze the deflection of a probe beam because of pump-probe interaction in a high-absorbing thermal medium. We extend the existing theory by accounting for translation of a finite-width probe because of deflection...
详细信息
We analyze the deflection of a probe beam because of pump-probe interaction in a high-absorbing thermal medium. We extend the existing theory by accounting for translation of a finite-width probe because of deflection within the nonlinear sample. We also provide expressions for the number of resolvable angles of the probe for possible applications of the setup as a beam deflector and study conditions for the maximization of the deflection angle and the resolution. We present experimental results obtained with a solution of chlorophyll in ethanol as the thermal medium.
The phase sensitivity with respect to coefficient errors is studied for a class of lattice digital allpass filters. It is shown that corresponding to each lattice parameter there is a set of frequencies for which the ...
详细信息
The phase sensitivity with respect to coefficient errors is studied for a class of lattice digital allpass filters. It is shown that corresponding to each lattice parameter there is a set of frequencies for which the phase response remains invariant to an arbitrary perturbation of the parameter. In addition, a simple result of sensitivity “amplification” in lossless two-pair networks is derived, which is used to obtain means and bounds on the sensitivity functions for these lattice structures in term of the lattice parameters. From these results some useful coefficient sensitivity properties are inferred. A practical example is included to better understand the interaction of sensitivity mechanisms in lattice allpass filter realizations, in the context of transfer functions realized as the parallel connection of two allpass filters. Finally, it is shown that the order reduction process intrinsic to the lattice synthesis procedure minimizes the L ∞ norm of the phase sensitivity at each step. Für eine Klasse digitaler Allpaßfilter in Gitterstruktur wird die Empfindlichkeit der Phase gergenüber Koeffizientenfehlern betrachtet. Es wird gezeigt, daß es bezüglich eines jeden Filterparameters gewisse Frequenzpunkte gibt, in denen der Phasengang auch bei beliebiger Parameterverfälschung unverändert bleibt. Darüber hinaus wird in einfacher Form eine “Empfindlichkeits-Verstärkung” in verlustfreien Signalfluß-Zweitor-Netzwerken hergeleitet. Sie wird dazu benützt, Mittel- und Grenzwerte für die Empfindlichkeitsfunktionen dieser Gitterstrukturen, ausgedrückt in den Filterparametern, zu gewinnen. Aus diesen Ergebnissen wird auf einige nützliche Eigenschaften der Koeffizientenempfindlichleit geschlossen. Ein praktisches Beispiel schließt sich an; es dient zum besseren Verständnis des Zusammenwirkens von Empfindlichkeitsmechanismen in Gitterstruktur-Allpässen, wie sie zur Realisierung von Übertrangungsfunktionen durch eine parallel-Anordnung von zwei Allpaßfiltern verwendet werden.
Plasma immersion ion implantation (PIII) is a technique which promises high dose rate implantation and compatibility with large-area processing. When a large negative bias is applied to the substrate which is immersed...
详细信息
Plasma immersion ion implantation (PIII) is a technique which promises high dose rate implantation and compatibility with large-area processing. When a large negative bias is applied to the substrate which is immersed inside a high ion-density plasma, all ion species present will be implanted without ion mass selection, Innovations of this techniques include: implantation time independent of implantation area, capability to perform concomitant deposition and implantation, and simplicity of machine design and maintenance. This paper reports the modeling of PIII plasma dynamics and several demonstrated semiconductor processing applications such as plasma doping, subsurface material synthesis, ion beam mixing, microcavity engineering, and surface modifications. Several processing issues such as substrate charging and dosimetry will also be discussed.
暂无评论