• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.34S no.4
• /
• pp.57-65
• /
• 1997

The purpose of this paper is to propose the HMM (hidden markov model) based on multi-observation sequence for the isolated word recognition. The proosed model generates the codebook of MSVQ by dividing each word into several sections followed by dividing training data into several sections. Then, we are to obtain the sequential value of multi-observation per each section by weighting the vectors of distance form lower values to higher ones. Thereafter, this the sequential with high probability value while in recognition. 146 DDD area names are selected as the vocabularies for the target recognition, and 10LPC cepstrum coefficients are used as the feature parameters. Besides the speech recognition experiments by way of the proposed model, for the comparison with it, the experiments by DP, MSVQ, and genral HMM are made with the same data under the same condition. The experiment results have shown that HMM based on multi-observation sequence proposed in this paper is proved superior to any other methods such as the ones using DP, MSVQ and general HMM models in recognition rate and time.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2000.11a
• /
• pp.28-31
• /
• 2000

We describe a hidden Markov model, HMMTIR, for general protein sequence based on the I-sites library of sequence-structure motifs. Unlike the linear HMMs used to model individual protein families, HMMSTR has a highly branched topology and captures recurrent local features of protein sequences and structures that transcend protein family boundaries. The model extends the I-sites library by describing the adjacencies of different sequence-structure motifs as observed in the database, and achieves a great reduction in parameters by representing overlapping motifs in a much more compact form. The HMM attributes a considerably higher probability to coding sequence than does an equivalent dipeptide model, predicts secondary structure with an accuracy of 74.6% and backbone torsion angles better than any previously reported method, and predicts the structural context of beta strands and turns with an accuracy that should be useful for tertiary structure prediction. HMMSTR has been incorporated into a public, fully-automated protein structure prediction server.

• Speech Sciences
• /
• v.9 no.4
• /
• pp.265-274
• /
• 2002

In this paper we propose an automatic segmentation system that outputs the time alignment information of phoneme boundary using Viterbi search with HMM (Hidden Markov Model) and corrects these results by an UVS (unvoiced/voiced/silence) classification algorithm. We selecte a set of 39 monophones and a set of 647 extended phones for HMM models. For the UVS classification we use the feature parameters such as ZCR (Zero Crossing Rate), log energy, spectral distribution. The result of forced alignment using the extended phone set is 11% better than that of the monophone set. The UVS classification algorithm shows high performance to correct the segmentation results.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2003.11a
• /
• pp.72-77
• /
• 2003

In this paper, an experimental verification of system identification technique for constructing finite element model is conducted for a three-story test structure equipped with an active mass driver (AMD). Twenty Gaussian white noises were used as the input for AMD, and the corresponding accelerations of each floors are measured. Then, the complex frequency response function (FRF) for the input, the force induced by the AMD, was obtained and subsequently, the Markov parameters and system matrices were estimated. The magnitudes as well as phase of experimentally obtained FRFs match well with those of analytically obtained FRFs.

• Journal of Ship and Ocean Technology
• /
• v.2 no.1
• /
• pp.42-57
• /
• 1998

Genetic Algorithm (GA) is applied to obtain optimal Inspection plan for fatigue deteriorating structures. The optimization problem is defined so as to minimize inspection cost in the 1ifs-time of the structure under the constraint that the increment of failure probability in each inspection interval is maintained below a target value. Optimization parameters are the inspection timing and the inspection quality. The inspection timing is selected from the discrete intervals such as one year, two years, three years, etc. The inspection quality is selected from the followings; no inspection, normal inspection, sampling inspection or precise inspection. The applicability of the proposed GA approach is demonstrated through the numerical calculations assuming a structure consisting of four member sets. Influences of the level of target failure probability, initial defect condition and stress increase due to plate thickness reduction caused by corrosion on inspection planning are discussed.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.39 no.2
• /
• pp.97-109
• /
• 2014

This paper considers a reverse supply chain with simultaneous recovery of used products and manufacturing of brand-new ones. Recovered products are downgraded and have to be sold in a market different from that of brand-new products at a different price. In case of a shortage of recovered product inventory, a brand-new item, if available, can be offered at the price of a recovered product. In other words, one-way demand substitution is allowed. We address the joint decision of when to manufacture brand-new product, when to recover returned product, and how to control demand substitution to maximize the hybrid production system's profits. To this end, we propose a Markov decision Process model and investigate the structure of the optimal policy. Performance comparison is numerically implemented between the models with and without downward demand substitution option under different operating conditions of the system parameters.

• Journal of the Korea Society for Simulation
• /
• v.9 no.2
• /
• pp.75-87
• /
• 2000

In this paper, nonstationary multiple sampling plans are discussed which are difficult to solve by analytical method when there exists dependency between the sample data. The initial solution is found by the sequential sampling plan using the sequential probability ration test. The number of acceptance and rejection in each step of the multiple sampling plan are found by grouping the sequential sampling plan's solution initially. The optimal multiple sampling plans are found by simulation. Four search methods are developed U and the optimum sampling plans satisfying the Type I and Type ll error probabilities. The performance of the sampling plans is measured and their algorithms are also shown. To consider the nonstationary property of the dependent sampling plan, simulation method is used for finding the lot rejection and acceptance probability function. As a numerical example Markov chain model is inspected. Effects of the dependency factor and search methods are compared to analyze the sampling results by changing their parameters.

• International Journal of Control, Automation, and Systems
• /
• v.2 no.3
• /
• pp.384-389
• /
• 2004

Recently a method of model reduction for discrete systems has been proposed in the literature based on a new impulse response Gramian. In this method, the system matrix$A_r$ of a reduced model is computed by approximating the reduced-order impulse response Gramian. The remaining matrices $b_r$ and $c_r$ are obtained so that various initial Markov parameters and time-moments of the original system are preserved in the reduced model. In this paper a different approach is presented based on the recursive relationship among the impulse response Gramians.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2002.05a
• /
• pp.134-138
• /
• 2002

The human faces do not have distinct features unlike other general objects. In general the features of eyes, nose and mouth which are first recognized when human being see the face are defined. These features have different characteristics depending on different human face. In this paper, We propose a face recognition algorithm using the hidden Markov model(HMM). In the preprocessing stage, we find edges of a face using the locally adaptive threshold scheme and extract features based on generic knowledge of a face, then construct a database with extracted features. In training stage, we generate HMM parameters for each person by using the forward-backward algorithm. In the recognition stage, we apply probability values calculated by the HMM to input data. Then the input face is recognized by the euclidean distance of face feature vector and the cross-correlation between the input image and the database image. Computer simulation shows that the proposed HMM algorithm gives higher recognition rate compared with conventional face recognition algorithms.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.28B no.10
• /
• pp.790-798
• /
• 1991

The congestion control methods of an integrated heterogeneous traffic multiplexer are studied. The real-time traffics have preemptive priority over nonreal-time traffics in capturing the common output link. Also real-time traffics are transmitted with bandwidth reduction when the nonreal-time traffics are over the buffer threshold. The proposed model formulated the system as a continous time Markov process and is analysed using matrix equation. Time delay and average number of used channel are applied for performance parameters. In this study, a new control method with the sizable buffer threshold is introduced and the optimal congestion control can be obtained.