• ETRI Journal
• /
• 제36권4호
• /
• pp.609-616
• /
• 2014

Even though many computational methods (recursive formulae) for blocking probabilities in finite-capacity M/D/1 queues have already been produced, these are forms of transforms or are limited to single-node queues. Using a distinctly different approach from the usual queueing theory, this study introduces explicit (transform-free) formulae for a blocking probability, a stationary probability, and mean sojourn time under either production or communication blocking policy. Additionally, the smallest buffer capacity subject to a given blocking probability can be determined numerically from these formulae. With proper selection of the overall offered load ${\rho}$, the approach described herein can be applicable to more general queues from a computational point of view if the explicit expressions of random vector $D_n$ are available.

• Communications for Statistical Applications and Methods
• /
• 제13권2호
• /
• pp.243-254
• /
• 2006

The relative frequency of order statistics is investigated for independent and identically distributed (i.i.d.) random variables. Specifically, it is shown that the probability $Pr\{X_{[s]}=x\}$ is no less than the probability $Pr\{X_{[r]}=x\}$ at any point $x{\geqq}x_0$ when r$X_{[r]}$ denotes the r-th order statistic of an i.i.d. discrete random vector and $x_0$ depends on the population probability distribution. A similar result for i.i.d. continuous random vectors is also presented.

• 말소리와 음성과학
• /
• 제2권1호
• /
• pp.77-85
• /
• 2010

This paper proposes an efficient feature vector processing technique to guard the Speech Emotion Recognition (SER) system against a variety of noises. In the proposed approach, emotional feature vectors are extracted from speech processed by comb filtering. Then, these extracts are used in a robust model construction based on feature vector classification. We modify conventional comb filtering by using speech presence probability to minimize drawbacks due to incorrect pitch estimation under background noise conditions. The modified comb filtering can correctly enhance the harmonics, which is an important factor used in SER. Feature vector classification technique categorizes feature vectors into either discriminative vectors or non-discriminative vectors based on a log-likelihood criterion. This method can successfully select the discriminative vectors while preserving correct emotional characteristics. Thus, robust emotion models can be constructed by only using such discriminative vectors. On SER experiment using an emotional speech corpus contaminated by various noises, our approach exhibited superior performance to the baseline system.

• Communications for Statistical Applications and Methods
• /
• 제26권1호
• /
• pp.35-46
• /
• 2019

Multivariate Confidence Region (MCR) cannot be used to obtain the confidence region of the mean vector of multivariate data when the normality assumption is not satisfied; however, the Quantile Confidence Region (QCR) could be used with a Multivariate Quantile Vector in these cases. The coverage rate of the QCR is better than MCR; however, it has a disadvantage because the QCR has a wide shape when the probability density function follows a bimodal form. In this study, we propose a Quantile Confidence Region using the Highest density (QCRHD) method with the Highest Density Region (HDR). The coverage rate of QCRHD was superior to MCR, but is found to be similar to QCR. The QCRHD is constructed as one region similar to QCR when the distance of the mean vector is close. When the distance of the mean vector is far, the QCR has one wide region, but the QCRHD has two smaller regions. Based on these features, it is found that the QCRHD can overcome the disadvantages of the QCR, which may have a wide shape.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.2594-2599
• /
• 2003

We consider some problems of the Modified SPRT(Sequential Probability Ratio Test) method for fault detection and isolation of inertial redundant sensor systems and propose an Advanced SPRT method to solve the problems of the Modified SPRT method. One problem of the Modified SPRT method to apply to inertial sensor system comes from the effect of inertial sensor errors such as misalignment, scale factor error and sensor bias in the parity vector, which make the Modified SPRT method hard to be applicable. The other problem is due to the correlation of parity vector components which may induce false alarm. We use a two-stage Kalman filter to remove effects of the inertial sensor errors and propose the modified parity vector and the controlled parity vector which removes the effect of correlation of parity vector components. The Advanced SPRT method is derived form the modified parity vector and the controlled parity vector. Some simulation results are presented to show the usefulness of the Advanced SPRT method to redundant inertial sensor systems.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 하계종합학술대회 논문집
• /
• pp.474-478
• /
• 1999

Genetic algorithm is an efficient search method using principles of natural selection and population genetics. In conventional genetic algorithms, however, the size of gene pool should be increased to insure a convergency. Therefore, many memory spaces and much computation time were needed. Also, since child chromosomes were generated by chromosome crossover and gene mutation, the algorithms have a complex structure. Thus, in this paper, a compact stereo matching algorithm using a population-based incremental teaming based on probability vector is proposed to reduce these problems. The PBIL method is modified for matching environment. Since the Proposed algorithm uses a probability vector and eliminates gene pool, chromosome crossover, and gene mutation, the matching algorithm is simple and the computation load is considerably reduced. Even if the characteristics of images are changed, stable outputs are obtained without the modification of the matching algorithm.

• 항공우주기술
• /
• 제4권1호
• /
• pp.66-76
• /
• 2005

이 연구는 통계학적인 방법을 이용하여 별 추적기의 가시도를 해석하였다. 별 추적기의 시선벡터에 대한 함수로서, 몇 가지 위성 자세 모드 상태에서 임의의 위성의 위치로부터 별 추적기가 태양과 지구에 의해 가려지는 확률, 태양 전지판에 의해 별 추적기가 가려지는 확률, 태양 전지판에 의한 태양 빛의 반사 확률을 계산하였다. 별 추적기의 이용도가 가장 높은 영역을 찾아 별 추적기의 시선벡터 방향을 결정하였는데, 두개의 최적화된 별추적기의 배치는 각각 고도각 -40도, 방위각 -35도와 고도각 -40도, 방위각 -150도로 나타났다.

• 말소리와 음성과학
• /
• 제2권3호
• /
• pp.141-148
• /
• 2010

In this paper, voice activity detection (VAD) for dual-channel noisy speech recognition is proposed in which spatial cues are employed. In the proposed method, a probability model for speech presence/absence is constructed using spatial cues obtained from dual-channel input signal, and a speech activity interval is detected through this probability model. In particular, spatial cues are composed of interaural time differences and interaural level differences of dual-channel speech signals, and the probability model for speech presence/absence is based on a Gaussian kernel density. In order to evaluate the performance of the proposed VAD method, speech recognition is performed for speech segments that only include speech intervals detected by the proposed VAD method. The performance of the proposed method is compared with those of several methods such as an SNR-based method, a direction of arrival (DOA) based method, and a phase vector based method. It is shown from the speech recognition experiments that the proposed method outperforms conventional methods by providing relative word error rates reductions of 11.68%, 41.92%, and 10.15% compared with SNR-based, DOA-based, and phase vector based method, respectively.

• 한국음향학회지
• /
• 제16권2호
• /
• pp.71-79
• /
• 1997

본 논문에서는 FVQ-DHMM(fuzzy vector quantization-discrete hidden Markov model)에서 강인한 출력확률의 추정을 위해서 코드워드 종속 거리 정규화와 출력확률에 대한 instar 형태의 퍼지 평활화 방법을 제안한다. FVQ-DHMM은 DHMM의 변형된 모델로, 상태별 출력확률이 입력패턴에 대한 각 코드워드와의 가중치와 출력확률의 곱에 대한 합의 형태로 추정된다. FVQ-DHMM의 성능이 가중치 요소와 상태별 출력분포에 영향을 받으므로, 가중치 요소와 상태별 출력분포를 강인하게 추정하는 방법이 필요하게 된다. 실험결과, 제안된 코드워드 종속 거리 정규화(CDDN : codeword dependent distance normalization)를 적용한 방법이 기존의 FVQ-DHMM에 비해 24%의 오인식률 감소가 있었으며, 상태별 출력분포에 대해서 평활화를 적용한 경우 79%의 오식율을 감소 시킴을 알 수 있었다. 이러한 결과는 제안된 CDDN과 퍼지 평활화의 사용이 향상된 인식율을 얻는데 주요하며, 결과적으로 제안된 방법이 FVQ-HMM을 위한 강인한 출력확률의 추정을 위한 대안으로 유용함을 보여준다고 할 수 있다.

• 정보처리학회논문지B
• /
• 제12B권4호
• /
• pp.429-436
• /
• 2005

본 논문에서는 유전자 알고리즘의 일반적인 문제점인 과도한 저장공간의 소모와 탐색의 비효율성을 줄이기 위해 PBIL을 이용한 단순한 스테레오 정합 기법을 제안한다. PBIL은 확률벡터에 기반해서 통계적 탐색과 경쟁학습을 이용하는 변종 유전자 알고리즘이며 확률벡터의 사용으로 인해 직렬 및 병렬 유전자 알고리즘군에 비해 단순한 구조를 가진다. 본 논문에서는 이 PBIL을 스테레오 정합 환경에 맞게 변형 및 단순화시켜 정합 알고리즘을 개발한다. 높은 적응성을 갖는 염색체는 생존 확률 또한 높다는 진화 법칙을 보존하면서 유전자 풀, 염색체 교차 및 유전자 돌연변이를 제거할 수 있으며 그 결과 저장공간을 줄이고 정합 규칙을 간소화하여 계산 비용을 감소시킬 수 있다. 추가적으로 다해상도 정합 기법처럼 넓은 영역의 변이 일관성을 획득하기 위해 변이 연속성에 대한 이웃들의 거리를 제어하는 방식을 추가하여 고정된 작은 정합창을 사용하면서 안정된 결과를 얻을 수 있게 한다. 마지막으로 단순한 시스템에 적용될 수 있게 하기 위해서 확률벡터를 사용하지 않는 제안한 알고리즘의 소형 대안 기법을 제시한다.