• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.5A
• /
• pp.355-362
• /
• 2009

In real-time communication services, delay constraints are among the most important QoS (Quality of Service) factors. In particular, it is difficult to guarantee the delay requirement over wireless channels, since they exhibit dynamic time-varying behavior and even severe burst-errors during periods of deep fading. Channel throughput may be increased, but at the cost of the additional delays when ARQ (Automatic Repeat Request) schemes are used. For real-time communication services, it is very essential to predict data deliverability. This paper derives the delay distribution and the successful delivery probability within a given delay budget using a priori channel model and a posteriori information from the perspective of queueing theory. The Gilbert-Elliot burst-noise channel is employed as an a Priori channel model, where a two-state Markov-modulated Bernoulli process $(MMBP_2)$ is used. for a posteriori information, the channel parameters, the queue-length and the initial channel state are assumed to be given. The numerical derivation is verified and analyzed via Monte Carlo simulations. This numerical derivation is then applied to a rate control scheme for real-time video transmission, where an optimal encoding rate is determined based on the future channel capacity and the distortion of the reconstructed pictures.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.3
• /
• pp.709-720
• /
• 2017

In healthcare and medical research, many important variables have a measurement error such as body mass index and laboratory data. It is also not easy to collect samples of large size because of high cost and long time required to collect the target patient satisfied with inclusion and exclusion criteria. Beside, the demand for solving a complex scientific problem has highly increased so that a semiparametric regression approach could be of substantial value solving this problem. To address the issues of measurement error, small domain and a scientific complexity, we conduct a multivariable Bayesian smoothing under structural measurement error covariate in this article. Specifically we enhance our previous model by incorporating other useful auxiliary covariates free of measurement error. For the regression spline, we use a radial basis functions with fixed knots for the measurement error covariate. We organize a fully Bayesian approach to fit the model and estimate parameters using Markov chain Monte Carlo. Simulation results represent that the method performs well. We illustrate the results using a national survey data for application.

• ETRI Journal
• /
• v.38 no.3
• /
• pp.444-454
• /
• 2016

Voice service is very demanding in cognitive radio networks (CRNs). The available spectrum in a CRN for CR users varies owing to the presence of licensed users. On the other hand, voice packets are delay sensitive and can tolerate a limited amount of delay. This makes the support of voice traffic in a CRN a complicated task that can be achieved by devising necessary considerations regarding the various network functionalities. In this paper, the support of secondary voice users in a CRN is investigated. First, a novel packet scheduling scheme that can provide the required quality of service (QoS) to voice users is proposed. The proposed scheme utilizes the maximum packet transmission rate for secondary voice users by assigning each secondary user the channel with the best level of quality. Furthermore, an analytical framework developed for a performance analysis of the system, is described in which the effect of erroneous spectrum sensing on the performance of secondary voice users is also taken into account. The QoS parameters of secondary voice users, which were obtained analytically, are also detailed. The analytical results were verified through the simulation, and will provide helpful insight in supporting voice services in a CRN.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.5
• /
• pp.24-31
• /
• 1994

In this paper, we study continuous speech recognition of Korean language using acoustic models of word juncture coarticulation. To alleviate the performance degradation due to coarticulation problems, we use context-dependent units that model inter-word transitions in addition to intra-word transitions. In all cases the initial phone of each word has to be specified for each possible final phone of the previous word similarly for the final phone of each word. To improve the robustness of the HMM parameters, the covariance matrix is smoothed. We also use position-dependent units to improve the discriminative power between units. Simulation results show that when the improved models of word juncture coarticulation are used. the recognition performance is considerably improved compared to the baseline system using only intra-word units.

• 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 Communications and Networks
• /
• v.4 no.2
• /
• pp.128-135
• /
• 2002

We propose a new space priority mechanism, and analyze its performance in a single Constant Bit Rate (CBR) server. The arrival process is derived from the superposition of two types of traffics, each in turn results from the superposition of homogeneous ON-OFF sources that can be approximated by means of a two-state Markov Modulated Poisson Process (MMPP). The buffer mechanism enables the Asynchronous Transfer Mode (ATM) layer to adapt the quality of the cell transfer to the Quality of Service (QoS) requirements and to improve the utilization of network resources. This is achieved by "Selective-Delaying and Pushing-ln"(SDPI) cells according to the class they belong to. The scheme is applicable to schedule delay-tolerant non-real time traffic and delay-sensitive real time traffic. Analytical expressions for various performance parameters and numerical results are obtained. Simulation results in term of cell loss probability conform with our numerical analysis.

• Journal of the Korea Society of Computer and Information
• /
• v.5 no.2
• /
• pp.90-97
• /
• 2000

This paper describes an speaker-adaptive speech recognition system which make a reliable recognition of speech signal for new speakers. In the Proposed method, an speech spectrum of new speaker is adapted to the reference speech spectrum by using Parameters of a 1st linear transformation network at the front of phoneme classification neural network. And the recognition system is based on semicontinuous HMM(hidden markov model) which use the multilayer perceptron as a fuzzy vector quantizer. The experiments on the isolated word recognition are performed to show the recognition rate of the recognition system. In the case of speaker adaptation recognition, the recognition rate show significant improvement for the unadapted recognition system.

• Proceedings of the IEEK Conference
• /
• summer
• /
• pp.92-96
• /
• 2004

The researches of the emotions are currently great interest in speech processing as well as in human-machine interaction domain. In the recent years, more and more of researches relating to emotion synthesis or emotion recognition are developed for the different purposes. Each approach uses its methods and its various parameters measured on the speech signal. In this paper, we proposed using a short-time parameter: MFCC coefficients (Mel­Frequency Cepstrum Coefficients) and a simple but efficient classifying method: Vector Quantification (VQ) for speaker-dependent emotion recognition. Many other features: energy, pitch, zero crossing, phonetic rate, LPC... and their derivatives are also tested and combined with MFCC coefficients in order to find the best combination. The other models: GMM and HMM (Discrete and Continuous Hidden Markov Model) are studied as well in the hope that the usage of continuous distribution and the temporal behaviour of this set of features will improve the quality of emotion recognition. The maximum accuracy recognizing five different emotions exceeds $88\%$ by using only MFCC coefficients with VQ model. This is a simple but efficient approach, the result is even much better than those obtained with the same database in human evaluation by listening and judging without returning permission nor comparison between sentences [8]; And this result is positively comparable with the other approaches.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.5
• /
• pp.1133-1146
• /
• 2016

Counts or averages over arbitrary regions are often analyzed using conditionally autoregressive (CAR) models. The spatial neighborhoods within CAR model are generally formed using only the inter-distance or boundaries between the sub-regions. Kyung and Ghosh (2009) proposed a new class of models to accommodate spatial variations that may depend on directions, using different weights given to neighbors in different directions. The proposed model, directional conditionally autoregressive (DCAR) model, generalized the usual CAR model by accounting for spatial anisotropy. Bayesian inference method is discussed based on efficient Markov chain Monte Carlo (MCMC) sampling of the posterior distributions of the parameters. The method is illustrated using a data set of median property prices across Greater Glasgow, Scotland, in 2008.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.1
• /
• pp.286-295
• /
• 2000

We present a novel statistically adaptive method using the Minimum Description Length(MDL) principle for unsupervised segmentation of magnetic resonance(MR) images. In the method, Markov random filed(MRF) modeling of tissue region accounts for random noise. Intensity measurements on the local region defined by a window are modeled by a finite Gaussian mixture, which accounts for image inhomogeneities. The segmentation algorithm is based on an iterative conditional modes(ICM) algorithm, approximately finds maximum ${\alpha}$ posteriori(MAP) estimation, and estimates model parameters on the local region. The size of the window for parameter estimation and segmentation is estimated from the image using the MDL principle. In the experiments, the technique well reflected image characteristic of the local region and showed better results than conventional methods in segmentation of MR images with inhomogeneities, especially.