• Speech Sciences
• /
• v.7 no.4
• /
• pp.11-26
• /
• 2000

In this paper, we propose the HMM with the MIN module. Because initial and re-estimated variance vectors are important elements for performance in HMM recognition systems, we propose a method which compensates for the mismatched statistical feature of training and test data. The MIN module function is a differentiable function similar to the sigmoid function. Unlike a continuous density function, it does not include variance vectors of the data set. The proposed hybrid HMM/MIN module is a unified network in which the observation probability in the HMM is replaced by the MIN module neural network. The parameters in the unified network are re-estimated by the gradient descent method for the Maximum Likelihood (ML) criterion. In estimating parameters, the variance vector is not estimated because there is no variance element in the MIN module function. The experiment was performed to compare the performance of the proposed HMM and the conventional HMM. The experiment measured an isolated number for speaker independent recognition.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.488-488
• /
• 2015

Intermittent streamflow is common phenomenon in arid and semi-arid regions. To manage water resources of intermittent streamflows, stochactic simulation data is essential; however the seasonally stochastic modeling for intermittent streamflow is a difficult task. In this study, using the periodic Markov chain model, we simulate intermittent monthly streamflow for occurrence and the periodic gamma autoregressive and copula models for amount. The copula models were tested in a previous study for the simulation of yearly streamflow, resulting in successful replication of the key and operational statistics of historical data; however, the copula models have never been tested on a monthly time scale. The intermittent models were applied to the Colorado River system in the present study. A few drawbacks of the PGAR model were identified, such as significant underestimation of minimum values on an aggregated yearly time scale and restrictions of the parameter boundaries. Conversely, the copula models do not present such drawbacks but show feasible reproduction of key and operational statistics. We concluded that the periodic Markov chain based the copula models is a practicable method to simulate intermittent monthly streamflow time series.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.3
• /
• pp.1034-1051
• /
• 2016

Mobility models are invaluable for determining the performance of routing protocols in opportunistic networks. The movement of nodes has a significant influence on the topological structure and data transmission in networks. In this paper, we propose a new mobility model called the campus-based community mobility model (CBCNM) that closely reflects the daily life pattern of students on a real campus. Consequent on a discovery that the pause time of nodes in their community follows a power law distribution, instead of a classical exponential distribution, we abstract the semi-Markov model from the movement of the campus nodes and analyze its rationality. Then, using the semi-Markov algorithm to switch the movement of the nodes between communities, we infer the steady-state probability of node distribution at random time points. We verified the proposed CBCNM via numerical simulations and compared all the parameters with real data in several aspects, including the nodes' contact and inter-contact times. The results obtained indicate that the CBCNM is highly adaptive to an actual campus scenario. Further, the model is shown to have better data transmission network performance than conventional models under various routing strategies.

• Korean Journal of Remote Sensing
• /
• v.24 no.1
• /
• pp.25-33
• /
• 2008

To investigate the hydrologic impacts of climate changes on dam inflow for Soyanggangdam watershed $(2694.4km^2)$ of northeastern South Korea, SLURP (Semi-distributed Land Use-based Runoff Process) model and the climate change results of CCCma CGCM2 based on SRES A2 and B2 were adopted. By the CA-Markov technique, future land use changes were estimated using the three land cover maps (1985, 1990, 2000) classified by Landsat TM satellite images. NDVI values for 2050 and 2100 land uses were estimated from the relationship of NDVI-Temperature linear regression derived from the observed data (1998-2002). Before the assessment, the SLURP model was calibrated and verified using 4 years (1998-2001) dam inflow data with the Nash-Sutcliffe efficiencies of 0.61 to 0.77. In case of A2 scenario, the dam inflows of 2050 and 2100 decreased 49.7 % and 25.0 % comparing with the dam inflow of 2000, and in case of B2 scenario, the dam inflows of 2050 and 2100 decreased 45.3 % and 53.0 %, respectively. The results showed that the impact of land use change covered 2.3 % to 4.9 % for the dam inflow change.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.13 no.3
• /
• pp.186-199
• /
• 2013

We pose pattern classification as a density estimation problem where we consider mixtures of generative models under partially labeled data setups. Unlike traditional approaches that estimate density everywhere in data space, we focus on the density along the decision boundary that can yield more discriminative models with superior classification performance. We extend our earlier work on the recursive estimation method for discriminative mixture models to semi-supervised learning setups where some of the data points lack class labels. Our model exploits the mixture structure in the functional gradient framework: it searches for the base mixture component model in a greedy fashion, maximizing the conditional class likelihoods for the labeled data and at the same time minimizing the uncertainty of class label prediction for unlabeled data points. The objective can be effectively imposed as individual mixture component learning on weighted data, hence our mixture learning typically becomes highly efficient for popular base generative models like Gaussians or hidden Markov models. Moreover, apart from the expectation-maximization algorithm, the proposed recursive estimation has several advantages including the lack of need for a pre-determined mixture order and robustness to the choice of initial parameters. We demonstrate the benefits of the proposed approach on a comprehensive set of evaluations consisting of diverse time-series classification problems in semi-supervised scenarios.

• Journal of the Korean Society of Safety
• /
• v.33 no.6
• /
• pp.37-41
• /
• 2018

Process safety technology has developed from qualitative methods such as HAZOP (hazard and operability study) to semi-quantitative methods such as LOPA (layer of protection analysis), and quantitative methods are actively studied these days. Quantitative risk assessment (QRA) is often based on fault tree analysis (FTA). FTA is efficient, but difficult to apply when failure events are not independent of each other. This problem can be avoided using a Markov process (MP). MP requires definition of all possible states, and thus, generally, is more complicated than FTA. A method is proposed in this work that uses an MP model and a Weibull distribution model in order to construct a reliability model for multiple dependent failures. As a case study, a pressure safety valve (PSV) is considered, for which there are three kinds of failure, i.e. open failure, close failure, and gas tight failure. According to recently reported inspection results, open failure and close failure are dependent on each other. A reliability model for a PSV group is proposed in this work that is to reproduce these results. It is expected that the application of the proposed method can be expanded to QRA of various systems that have partially dependent multiple failure states.

• ETRI Journal
• /
• v.28 no.3
• /
• pp.301-310
• /
• 2006

In mobile communications, a class of variable-complexity algorithms for convolutional decoding known as sequential decoding algorithms is of interest since they have a computational time that could vary with changing channel conditions. The Fano algorithm is one well-known version of a sequential decoding algorithm. Since the decoding time of a Fano decoder follows the Pareto distribution, which is a heavy-tailed distribution parameterized by the channel signal-to-noise ratio (SNR), buffers are required to absorb the variable decoding delays of Fano decoders. Furthermore, since the decoding time drawn by a certain Pareto distribution can become unbounded, a maximum limit is often employed by a practical decoder to limit the worst-case decoding time. In this paper, we investigate the relations between buffer occupancy, decoding time, and channel conditions in a system where the Fano decoder is not allowed to run with unbounded decoding time. A timeout limit is thus imposed so that the decoding will be terminated if the decoding time reaches the limit. We use discrete-time semi-Markov models to describe such a Fano decoding system with timeout limits. Our queuing analysis provides expressions characterizing the average buffer occupancy as a function of channel conditions and timeout limits. Both numerical and simulation results are provided to validate the analytical results.

• Convergence Security Journal
• /
• v.8 no.1
• /
• pp.91-100
• /
• 2008

The previous works in sensor networks security have focused on the aspect of confidentiality, authentication and integrity based on cryptographic primitives. There has been no prior work to assess the survivability in systematic way. Accordingly, this paper presents a survivability model of wireless sensor networks using software rejuvenation for dual adaptive cluster head. The survivability model has state transition to reflect status of real wireless sensor networks. In this paper, we only focus on a survivability model which is capable of describing cluster head compromise in the networks and able to switch over the redundant cluster head in order to increase the survivability of that cluster. Second, this paper presents how to enhance the survivability of sensor networks using software rejuvenation methodology for dual cluster head in wireless sensor network. We model and analyze each cluster as a stochastic process based on Semi Markov Process (SMP) and Discrete Time Markov Chain (DTMC). The proof of example scenarios and numerical analysis shows the feasibility of our approach.

• KIISE Transactions on Computing Practices
• /
• v.21 no.1
• /
• pp.94-99
• /
• 2015

With the recent machine learning paradigm of using nonparametric Bayesian statistics and statistical inference based on random sampling, the Dirichlet distribution finds many uses in a variety of graphical models. It is a multivariate generalization of the gamma distribution and is defined on a continuous (K-1)-simplex. This paper presents a sampling method for a Dirichlet distribution for the problem of dividing an integer X into a sequence of K integers which sum to X. The target samples in our problem are all positive integer vectors when multiplied by a given X. They must be sampled from the correspondingly gridded simplex. In this paper we develop a Markov Chain Monte Carlo (MCMC) proposal distribution for the neighborhood grid points on the simplex and then present the complete algorithm based on the Metropolis-Hastings algorithm. The proposed algorithm can be used for the Markov model, HMM, and Semi-Markov model for accurate state-duration modeling. It can also be used for the Gamma-Dirichlet HMM to model q the global-local duration distributions.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.11 no.1
• /
• pp.7-11
• /
• 1986

Throughout this paper we analyze the system at output point t of two stage cyclic queueing model. Our main result characterize the stochastic process (X$^{o}$ , T$^{o}$ ), the system at output point, as a Markov renewal process. The subsequent lemma exhibits the semi-Markov kernel of (X$^{o}$ , T$^{o}$ ) with state dependent feedback, the possibility of a reducible state space arises. A simple necessary and sufficient condition for the irreducibility of (X$^{o}$ , T$^{o}$ was determinded. This irreducibility implied that (X$^{o}$ , T$^{o}$ ) was aperiodic.