• 전자공학회논문지B
• /
• 제31B권8호
• /
• pp.11-19
• /
• 1994

The matching probability P(ο/$\lambda$), of the signal sequence(ο) observed for a finite time interval with a HMM (Hidden Markov Model $\lambda$) indicates the probability that signal comes from the given model. By utilizing the fact that the probability represents matching score of the observed signal with the model we can recognize an unknown signal pattern by comparing the magnitudes of the matching probabilities with respect to the known models. Because the algorithm however uses floating point variables during the computing process hardware implementation of the algorithm requires floating point units. This paper proposes an integer algorithm which uses positive integer numbers rather than float point ones to compute the matching probability so that we can economically realize the algorithm into hardware. The algorithm makes the model parameters integer numbers by multiplying positive constants and prevents from divergence of data through the normalization of variables at each step. The final equation of matching probability is composed of constant terms and a variable term which contains logarithm operations. A scheme to make the log conversion table smaller is also presented. To analyze the qualitive characteristics of the proposed algorithm we attatch simulation result performed on two groups of 10 hypothetic models respectively and inspect the statistical properties with repect to the model order the magnitude of scaling constants and the effect of the observation length.

• 한국해안해양공학회:학술대회논문집
• /
• 한국해안해양공학회 2003년도 한국해안해양공학발표논문집
• /
• pp.376-381
• /
• 2003

The reliability analysis of overtopping probability is proposed. In order to estimate the expected overtopping probability of the rubble mound breakwater, the experimental results of individual wave runup height is applied for the analysis of irregular wave system. The joint distribution of wave heights and periods is used for the input data of runup calculation because the runup height depends on the wave height and period. The runup heights during the one event that the design wave attacks the rubble mound breakwater extend to the one life cycle of 60 years. Utilizing the Monte-Carlo method, the one life cycle is tried more about 60 times for obtaining the expected value of overtopping probability. It is found that the inclusion of the variability of wave tidal and wave steepness has great influence on the computation of the expected overtopping probability of rubble mound breakwater. The previous design disregarding the tidal fluctuation largely overestimates or underestimates the expected overtopping probability depending on tidal range and wave steepness.

• 전자공학회논문지B
• /
• 제31B권7호
• /
• pp.211-218
• /
• 1994

Stochastic computation employs random pulse streams to represent numbers. In this paper, we study a new method to implement the number system which uses the ratio of the numbers of ones and zeros in the pulse streams. In this number system. if P is the probability that a pulse is one in a pulse stream then the number X represented by the pulse stream is defined as P/(1-P). We propose circuits to implement the basic operations such as addition multiplication and sigmoid function with this number system and examine the error characteristics of such operations in stochastic computation. We also propose a neuron model and derive a learning algorithm based on backpropagation for the 3-layered feedforward neural networks. We apply this learning algorithm to a digit recognition problem. To analyze the results, we discuss the errors due to the variance of the random pulse streams and the quantization noise of finite length register.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2004년도 학술발표회
• /
• pp.479-483
• /
• 2004

A new calculation method of expected overtopping probability of rubble mound breakwater considering real tide occurrence has been proposed. A calculation method of expected overtopping probability of rubble mound breakwater was proposed by Kweon and Suh (2003). In their calculation, the fluctuation of tidal elevation was expressed by the sinusoidal change that yields the uniform distribution of occurrence frequency. However, the realistic distribution of tidal elevation should influence on the overtopping chance. In this study, the occurrence frequency of tidal elevation obtained from the real sea is included. The tidal elevation used in this study is collected from the east coastal part of Korean peninsular. Analyzing the annual data of the tidal fluctuation measured hourly during 355 days, the distribution of occurrence frequency is formulated utilizing by the normal distribution with one peak. Among the calculation procedures of annual maximum wave height, wave height-period joint distribution, wave run-up height and occurrence frequency of tide, only the annual maximum wave height is again chosen randomly from normal distribution to consider the uncertainty. The others are treated by utilizing the distribution function or relationship itself, It is found that the inclusion of the variability of tidal elevation has great influence on the computation of the expected overtopping probability of rubble mound breakwater. The bigger standard deviation of occurrence frequency is, the lower the overtopping probability of rubble mound breakwater is.

• 한국군사과학기술학회지
• /
• 제15권5호
• /
• pp.601-614
• /
• 2012

Since the optimized use of sonar systems available for detection is a very practical problem for a given ocean environment, the measure of mission achievability is needed for operating the sonar system efficiently. In this paper, a theory on Measure Of Effectiveness(MOE) for specific mission such as detection is described as the measure of mission achievability, and a recursive Cumulative Detection Probability(CDP) algorithm is found to be most efficient from comparing three CDP algorithms for discrete glimpses search to reduce computation time and memory for complicated scenarios. The three CDPs which are MOE for sonar-maneuver pattern are calculated as time evolves for comparison, based on three different formula depending on the assumptions as follows; dependent or independent glimpses, unimodal or non-unimodal distribution of Probability of Detection(PD) as a function of observation time interval for detection. The proposed CDP algorithm which is made from unimodal formula is verified and applied to OASPP(Optimal Acoustic Search Path Planning) with complicated scenarios.

• 한국안전학회지
• /
• 제24권4호
• /
• pp.53-58
• /
• 2009

The importance of process for repair and reinforcement of the bridge is increasing because of the lack of the fatigue load and stress, a lowering of the bridge load carrying capacity owing to impact and oscillation, deterioration on cultivation periods of the bridge, etc. Typically the experimenter values the bridge load carrying capacity by the real rating factor and response modification factor in bridge load rating through static load test and dynamic load test. But the error occurred in reliability of response modification factor in bridge load rating according to experience of experimenter. so tests of connecting probability theory and valuation of the bridge recently. The study is to compute the real load carrying capacity of the bridge and the rating factor and response modification factor on grade of the bridge, and calculate the probability of over-loaded truck load from Weigh In Motion(WIM) Data in FORTRAN programming applying to Monte-Carlo Simulation. At the result of this study, it is acquired that the new grade is computed for the probability of over-loaded truck load and surface inspection. The A grade is over 1.95, B grade is $1.55{\sim}1.94$, C grade is $1.26{\sim}1.54$, D grade is $1.14{\sim}1.25$, E grade is under 1.13 of rating factor, respectively.

• 한국정보과학회논문지:정보통신
• /
• 제29권2호
• /
• pp.181-187
• /
• 2002

Connection admission control(CAC), which decides whether or not to accept a new call request, is one of the most Important preventive congestion control techniques in asynchronous transfer mode(ATM) networks. To develop a practical CAC scheme, first we propose a "Modified Cell Loss Probability MP${\nu}"$, which is based on "Virtual Cell Loss Probability P${\nu}"$, taking into account mean burst duration of input traffic source and buffer size in ATM networks. MP${\nu}"$ computes more accurate cell loss probability than P${\nu}"$ without increasing computational complexity, since P${\nu}"$ is formulated simply form the maximum and the average cell rate of input traffic. P${\nu}"$ is overestimated as compared to the real cell loss probability when the mean burst duration is relatively small to the buffer capacity. Then, we Propose a CAC scheme, based on "Modified Virtual Bandwidth(MVB)" method, which may individualize the cell loss probabilities in heterogeneous traffic environments. For the proposed approach, we define the interference intensity to identify interferences between heterogeneous traffic sources and use it as well as MP${\nu}"$ to compute MVB. Our approach is well suitable for ATM networks since it provides high bandwidth utilization and guarantees simple and real time CAC computation for heterogeneous traffic environments.heterogeneous traffic environments.

• Structural Engineering and Mechanics
• /
• 제78권2호
• /
• pp.209-218
• /
• 2021

For engineering, there are two major challenges in reliability analysis. First, to ensure the accuracy of simulation results, mechanical products are usually defined implicitly by complex numerical models that require time-consuming. Second, the mechanical products are fortunately designed with a large safety margin, which leads to a low failure probability. This paper proposes an efficient and high-precision adaptive active learning algorithm based on the Kriging surrogate model to deal with the problems with low failure probability and time-consuming numerical models. In order to solve the problem with multiple failure regions, the adaptive kernel-density estimation is introduced and improved. Meanwhile, a new criterion for selecting points based on the current Kriging model is proposed to improve the computational efficiency. The criterion for choosing the best sampling points considers not only the probability of misjudging the sign of the response value at a point by the Kriging model but also the distribution information at that point. In order to prevent the distance between the selected training points from too close, the correlation between training points is limited to avoid information redundancy and improve the computation efficiency of the algorithm. Finally, the efficiency and accuracy of the proposed method are verified compared with other algorithms through two academic examples and one engineering application.

• 한국군사과학기술학회지
• /
• 제13권5호
• /
• pp.824-830
• /
• 2010

It is not easy job to find a underwater target using sonar system in the ASW operations. Many researchers have tried to solve anti-submarine search problem aiming to maximize the probability of detection under limited searching conditions. The classical 'Search Theory' deals with search allocation problem and search path problem. In both problems, the main issue is to prioritize the searching cells in a searching area. The number of possible searching path that is combination of the consecutive searching cells increases rapidly by exponential function in the case that the number of searching cells or searchers increases. The more searching path we consider, the longer time we calculate. In this study, an effective algorithm that can maximize the probability of detection in shorter computation time is presented. We show the presented algorithm is quicker method than previous algorithms to solve search problem through the comparison of the CPU computation time.

• International Journal of Control, Automation, and Systems
• /
• 제6권1호
• /
• pp.109-118
• /
• 2008

We present two estimators for discrete non-Gaussian and nonstationary probability density estimation based on a dynamic Bayesian network (DBN). The first estimator is for off line computation and consists of a DBN whose transition distribution is represented in terms of kernel functions. The estimator parameters are the weights and shifts of the kernel functions. The parameters are determined through a recursive learning algorithm using maximum likelihood (ML) estimation. The second estimator is a DBN whose parameters form the transition probabilities. We use an asymptotically convergent, recursive, on-line algorithm to update the parameters using observation data. The DBN calculates the state probabilities using the estimated parameters. We provide examples that demonstrate the usefulness and simplicity of the two proposed estimators.