Wyniki wyszukiwania - BazTech

1

Diagnosis of incipient faults in nonlinear analog circuits

Deng Y., Shi Y., Zhang W.

Metrology and Measurement Systems

|

2012

|

Vol. 19, nr 2

203-218

EN

Considering the problem to diagnose incipient faults in nonlinear analog circuits, a novel approach based on fractional correlation is proposed and the application of the subband Volterra series is used in this paper. Firstly, the subband Volterra series is calculated from the input and output sequences of the circuit under test (CUT). Then the fractional correlation functions between the fault-free case and the incipient faulty cases of the CUT are derived. Using the feature vectors extracted from the fractional correlation functions, the hidden Markov model (HMM) is trained. Finally, the well-trained HMM is used to accomplish the incipient fault diagnosis. The simulations illustrate the proposed method and show its effectiveness in the incipient fault recognition capability.

2

Metoda filtru cząsteczkowego

Brzozowska-Rup K., Dawidowicz A.L.

Matematyka Stosowana : matematyka dla społeczeństwa

|

2009

|

T. 10/51

69-107

EN

This paper provides an introduction to the field of sequential Monte Carlo methods which are also known as particle filters methods. The best known algorithm to solve the problem of non-linear non-Gaussian filtering is the Extended Kalman Filter (EKF) but in settings where the dynamics are significantly non-linear or the noise intensities are high, the EKF can perform quite poorly. Particle filtering methods are powerful tools for online estimation and tracking in nonlinear and non-Gaussian dynamical systems. The basic idea is to approximate the probability density function [...] by a discrete cloud of points [...] called particles. They commonly consist of three steps: (1) drawing samples in the state-space of the system, (2) computing proper importance weights of each sample and (3) resampling. These methods are becoming increasingly popular in economics and finance so the objective of this paper is to explain the basic assumptions of the methodology and provide references to relevant literature.

PL

Celem pracy jest zaprezentowanie idei metod filtracji opartych na sekwencyjnej metodzie Monte Carlo, w literaturze nazywanych również metodami filtru cząsteczkowego oraz odniesienia do odpowiedniej literatury. Istotęo mawianych algorytmów przedstawiamy rozważając problem estymacji silnie nieliniowych i niegaussowskich modeli przestrzeni stanów. W praktyce bowiem w takich przypadkach dobrze znany i najczęściej wykorzystywany algorytm rozszerzonego filtru Kalmana (ang. Extended Kalman Filter, EKF) może wykazywać istotną nieefektywność. Idea filtru cząsteczkowego polega na estymacji rozkładu prawdopodobieństwa rozkładem empirycznym wyznaczonym na podstawie cząsteczek, tzw. "chmury punktów". Zaimplementowanie algorytmu filtru cząsteczkowego wymaga zasadniczo przeprowadzenia trzech procedur: (1) losowania cząsteczek z odpowiednio dobranej sekwencyjnej funkcji ważności, (2) określenia istotności cząsteczek, (3) powtórnego losowania, tzw. resampling. Metody te są coraz bardziej popularne w dziedzinie ekonomii, statystyki, medycynie i teorii sygnałów.

3

Genetic Algorithms as an Alternative Method of Parameter Estimation and Finding Most Likely Sequences of States of Hidden Markov Chains for HMMs and Hybrid HMM/ANN Models

Bijak K.

Fundamenta Informaticae

|

2008

|

Vol. 86, nr 1-2

1-17

EN

In this paper genetic algorithms are used in estimation and decoding processes of a Hidden Markov Model (HMM) and a hybrid HMM/ANN model with conditional binomial distributions. The hybrid model combines a hidden Markov chain with a perceptron which is assumed to constitute a match network. Genetic algorithms are applied here instead of the traditional methods such as the EM algorithm and the Viterbi algorithm. The paper demonstrates performance of an HMM and a hybrid model in modeling the annual number of months, in which some seismic events are recorded. Parameters of the discrete-time two-state models are estimated using the maximum likelihood method, on the basis of data on seismic events that were recorded in Romania in years 1901C1990. Then, on the basis of the estimation results, the most likely sequences of states of the hidden Markov chains are found.

4

Hierarchical Hidden Markov Models for User/Process Profile Learning

Galassi U., Botta M., Giordana A.

Fundamenta Informaticae

|

2007

|

Vol. 78, nr 4

487-505

EN

This paper presents an algorithm for automatically constructing sophisticated user/process profiles from traces of their behavior. A profile is encoded by means of a Hierarchical Hidden Markov Model (HHMM), which is a well formalized tool suitable to model complex patterns in long temporal or spatial sequences. A special sub-class of this hierarchical model, oriented to user/process profiling, is also introduced. The algorithm follows a bottom-up strategy, in which elementary facts in the sequences (motifs) are progressively grouped, thus building the abstraction hierarchy of a HHMM, layer after layer. The method is firstly evaluated on artificial data. Then a user identification task, from real traces, is considered. A preliminary experimentation with several different users produced encouraging results.