• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1995.10b
• /
• pp.233-242
• /
• 1995

In this paper, we introduce the identification model of dynamic system using the neural networks, We propose two identification models. The output of the parallel identification model is a linear combination of its past values as well as those of the input. The series-parallel model is a linear combination of the past values in the input and output of the plant. To generate stable adaptive laws, we prove that the series-parallel model is found to be proferable.

• The Journal of the Acoustical Society of Korea
• /
• v.32 no.3
• /
• pp.252-261
• /
• 2013

The common speech recognition system displays higher recognition performance in a quiet environment, while its performance declines sharply in a real environment where there are noises. To implement a speech recognizer that is robust in different speech settings, this study suggests the method of Parallel Model Combination adaptation using frequency-variant based on environment-awareness (FV-PMC), which uses variants in frequency; acquires the environmental data for speech recognition; applies it to upgrading the speech recognition model; and promotes its performance enhancement. This FV-PMC performs the speech recognition with the recognition model which is generated as followings: i) calculating the average frequency variant in advance among the readily-classified noise groups and setting it as a threshold value; ii) recalculating the frequency variant among noise groups when speech with unknown noises are input; iii) regarding the speech higher than the threshold value of the relevant group as the speech including the noise of its group; and iv) using the speech that includes this noise group. When noises were classified with the proposed FV-PMC, the average accuracy of classification was 56%, and the results from the speech recognition experiments showed the average recognition rate of Set A was 79.05%, the rate of Set B 79.43%m, and the rate of Set C 83.37% respectively. The grand mean of recognition rate was 80.62%, which demonstrates 5.69% more improved effects than the recognition rate of 74.93% of the existing Parallel Model Combination with a clear model, meaning that the proposed method is effective.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.7
• /
• pp.603-611
• /
• 2003

This paper proposes an effective feature compensation scheme based on speech model for achieving robust speech recognition. Conventional model-based method requires off-line training with noisy speech database and is not suitable for online adaptation. In the proposed scheme, we can relax the off-line training with noisy speech database by employing the parallel model combination technique for estimation of correction factors. Applying the model combination process over to the mixture model alone as opposed to entire HMM makes the online model combination possible. Exploiting the availability of noise model from off-line sources, we accomplish the online adaptation via MAP (Maximum A Posteriori) estimation. In addition, the online channel estimation procedure is induced within the proposed framework. For more efficient implementation, we propose a selective model combination which leads to reduction or the computational complexities. The representative experimental results indicate that the suggested algorithm is effective in realizing robust speech recognition under the combined adverse conditions of additive background noise and channel distortion.

• Particle and aerosol research
• /
• v.8 no.2
• /
• pp.75-81
• /
• 2012

The strength of lumped cokes can be represented by some index numbers. Although some indexes are suggested, these indexes are not enough to enlighten fracture mechanism. To find essential mechanism, a computational way, discrete element method, is applied to the uniaxial compression test for cylindrical specimen. The cylindrical specimen is a kind of lumped particle mass with parallel bonding that will be broken when the normal stress and shear stress is over a critical value. It is revealed that the primary factors for cokes fracture are parallel spring constant, parallel bond strength, bonding radius and packing ratio the parallel bond strength and radius of the parallel combination the packing density. Especially, parallel spring constant is directly related with elastic constant and yield strength.

• International Journal of Control, Automation, and Systems
• /
• v.5 no.5
• /
• pp.559-569
• /
• 2007

In this paper, the dynamic controller design problem of a redundant planar 2-dof parallel manipulator is studied. Using the Euler-Lagrange equation, we formulate the dynamic model of the parallel manipulator in the joint space and propose an augmented PD controller with forward dynamic compensation for the parallel manipulator. By formulating the controller in the joint space, we eliminate the complex computation of the Jacobian matrix of joint angles with end-effector coordinate. So with less computation, our controller is easier to implement, and a shorter sampling period can be achieved, which makes the controller more suitable for high-speed motion control. Furthermore, with the combination of static friction model and viscous friction model, the active joint friction of the parallel manipulator is studied and compensated in the controller. Based on the dynamic parameters of the parallel manipulator evaluated by direct measurement and identification, motion control experiments are implemented. With the experiments, the validity of the dynamic model is proved and the performance of the controller is evaluated. Experiment results show that, with forward dynamic compensation, the augmented PD controller can improve the tracking performance of the parallel manipulator over the simple PD controller.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.2
• /
• pp.21-26
• /
• 2001

An expectation-maximization (EM) based Bayesian adaptation method for the mean of noise is proposed for noise-robust speech recognition. In the algorithm, the on-line testing utterances are used for the unsupervised Bayesian adaptation and the prior distribution of the noise mean is estimated using the off-line training data. For the noisy speech modeling, the parallel model combination (PMC) method is employed. The proposed method has shown to be effective compared with the conventional PMC method for the speech recognition experiments in a car-noise condition.

• Journal of Institute of Control, Robotics and Systems
• /
• v.13 no.6
• /
• pp.560-567
• /
• 2007

The objective of this research is to optimize the designing parameters of the parallel manipulator with large orientation workspace at the boundary position of the constant orientation workspace (COW). The method uses a simple genetic algorithm(SGA) while considering three different kinematic performance indices: COW and the global conditioning index(GCI) to evaluate the mechanism's dexterity for translational motion of an end-effector, and orientation workspace of two angle of Euler angles to obtain the large rotation angle of an end-effector at the boundary position of COW. Total fifteen cases divided according to the combination of the sphere radius of COW and rotation angle of orientation workspace are studied, and to decide the best model in the total optimized cases, the fuzzy inference system is used for each case's results. An optimized model is selected as a best model, which shows better kinematic performances compared to the basis of the pre-existing model.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.2E
• /
• pp.25-32
• /
• 1995

In this paper, we propose to use a parallel-branch subunit model for improved word recognition. The model is obtained by splitting off each subunit branch based on mixture component in continuous hidden Markov model(continuous HMM). According to simulation results, the proposed model yields higher recognition rate than the single-branch subunit model or the parallel-branch subunit model proposed by Rabiner et al[1]. We show that a proper combination of the number of mixture components and the number of branches for each subunit results in increased recognition rate. To study the recognition performance of the proposed algorithms, the speech material used in this work was a vocabulary with 1036 Korean words.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.4
• /
• pp.959-968
• /
• 1999

several papers have reported that parallel heuristics or hybrid approaches combining several heuristics can get better results. However, the parallelization and hybridization of any search methods on the single CPU type computer need enormous computation time. that case, we need more elegant combination method. For this purpose, we propose Fast Model of Parallel Heuristics(FMPH). FMPH is based on the island model of parallel genetic algorithms and takes local search to the elite solution obtained form each island(sub group). In this paper we introduce how can we adapt FMPH to the job-shop scheduling problem notorious as the most difficult NP-hard problem and report the excellent results of several famous benchmark problems.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.14 no.4
• /
• pp.391-406
• /
• 2014

The performance of General-Purpose computation on Graphics Processing Units (GPGPU) is heavily dependent on the memory access behavior. This sensitivity is due to a combination of the underlying Massively Parallel Processing (MPP) execution model present on GPUs and the lack of architectural support to handle irregular memory access patterns. Application performance can be significantly improved by applying memory-access-pattern-aware optimizations that can exploit knowledge of the characteristics of each access pattern. In this paper, we present an algorithmic methodology to semi-automatically find the best mapping of memory accesses present in serial loop nest to underlying data-parallel architectures based on a comprehensive static memory access pattern analysis. To that end we present a simple, yet powerful, mathematical model that captures all memory access pattern information present in serial data-parallel loop nests. We then show how this model is used in practice to select the most appropriate memory space for data and to search for an appropriate thread mapping and work group size from a large design space. To evaluate the effectiveness of our methodology, we report on execution speedup using selected benchmark kernels that cover a wide range of memory access patterns commonly found in GPGPU workloads. Our experimental results are reported using the industry standard heterogeneous programming language, OpenCL, targeting the NVIDIA GT200 architecture.