• ETRI Journal
• /
• 제35권1호
• /
• pp.100-108
• /
• 2013

In this paper, a feature extraction (FE) method is proposed that is comparable to the traditional FE methods used in automatic speech recognition systems. Unlike the conventional spectral-based FE methods, the proposed method evaluates the similarities between an embedded speech signal and a set of predefined speech attractor models in the reconstructed phase space (RPS) domain. In the first step, a set of Gaussian mixture models is trained to represent the speech attractors in the RPS. Next, for a new input speech frame, a posterior-probability-based feature vector is evaluated, which represents the similarity between the embedded frame and the learned speech attractors. We conduct experiments for a speech recognition task utilizing a toolkit based on hidden Markov models, over FARSDAT, a well-known Persian speech corpus. Through the proposed FE method, we gain 3.11% absolute phoneme error rate improvement in comparison to the baseline system, which exploits the mel-frequency cepstral coefficient FE method.

• Journal of Electrical Engineering and Technology
• /
• 제10권4호
• /
• pp.1682-1691
• /
• 2015

In this paper, a real time implementation of selective harmonic elimination pulse width modulation (SHEPWM) using Real Coded Genetic Algorithm (RGA), Particle Swarm Optimization technique (PSO) and a new technique known as Linearization Method (LM) for Single Phase Matrix Converter (SPMC) is designed and discussed. In the proposed technique, the switching frequency is fixed and the optimum switching angles are obtained using simple mathematical calculations. A MATLAB simulation was carried out, and FFT analysis of the simulated output voltage waveform confirms the effectiveness of the proposed method. An experimental setup was also developed, and the switching angles and firing pulses are generated using Field Programmable Gate Array (FPGA) processor. The proposed method proves that it is much applicable in the industrial applications by virtue of its suitability in real time applications.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2006년도 춘계학술대회 논문집
• /
• pp.201-202
• /
• 2006

To analyze the dynamic characteristics of the bundle drawing process, we employed a Random Phase Spectrum method to generate stochastic test signals that had a given autocorrelation function. And the spectra of the dynamics of the process outputs were obtained, based on the dynamic model of the bundle drawing process. Results showed that the RPS method was very effective to generate stochastic signals that had an exponential function form. The drawing process had the traits that there existed a special frequency range, incurring the process resonance.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2014년도 추계학술대회 논문집
• /
• pp.555-556
• /
• 2014

In this paper, topology optimization of two-dimensional acoustic lenses is presented by using the phase field method. The objective of the optimization is to maximize the acoustic pressure at a specified domain inside the acoustic domain for a given frequency, and the constraint is imposed on the amount of the material of the acoustic lens. Topology optimization of two-dimensional acoustic lenses are obtained as the steady state of the phase transition described by the Allen-Cahn equation. The Helmholtz equation modeling the wave propagation is solved by using a finite element method. The effectiveness of the proposed method is verified by applying it for several two-dimensional acoustic lens system design problems.

• 대한전기학회논문지:전기기기및에너지변환시스템부문B
• /
• 제53권9호
• /
• pp.562-568
• /
• 2004

The frequency and phase angle of the utility voltage are important in many industrial systems. In the three-phase system, they can be easily known by using the utility voltage vector. However, in the case of single phase system, there are some difficulties in detecting the information of utility voltage. In conventional system, the zero-crossing detection method is widely used, but could not obtain the information of utility voltage instantaneously. In this paper, the new digital PLL control using virtual two phase detector is proposed with a detailed analysis of single-phase digital PLL control for utility connected systems. The experimental results under various utility conditions are presented and demonstrate an excellent phase tracking capability in the single-phase grid-connected operation.

• 한국통신학회논문지
• /
• 제32권7C호
• /
• pp.658-666
• /
• 2007

일반적으로 기존의 선형 위상을 갖는 저역통과 MAXFLAT FIR 필터 설계 방법들은 폐쇄형 전달함수의 특성으로 인하여 하프 주파수 대역 $(0{\leq}w{\leq}{\pi}/2)$에서 이득 응답을 갖는다. 더욱이 주파수 영역에서 MAXFLAT과 선형 위상 특성에 도달하기 위해 추정 알고리즘을 사용하기 때문에 설계 필터들은 저지대역에서의 리플 에러, 완만한 차단감쇠, 위상 및 그룹지연 그리고 부정확한 차단주파수와 같은 문제점들을 가지고 있다. 본 논문에서는 이러한 문제점들을 해결하기 위하여 제로 위상의 광대역 이득 응답을 갖는 MAXFLAT FIR 저역통과 필터 설계를 위한 새로운 수학적인 폐쇄형 전달함수를 제안하였다. 더욱이, 제안된 폐쇄형 전달함수는 임의의 차단 주파수를 포함한 MAXFLAT 조건들을 이용하여 새롭게 유도한 일반화한 공식들로 쉽게 실현됨이 증명되었다. 따라서 제안방식은 간단하고 빠른 설계에 적합하다. 결론적으로 제안된 설계 기술은 제로 위상과 광대역 이득 응답을 갖는 MAXFLAT FIR 저역통과 필터 실현이 가능하며, 설계된 필터들은 모두 250dB가 넘는 최대 차단 감쇠를 갖는 것을 알 수가 있다.

• 한국광학회지
• /
• 제17권6호
• /
• pp.507-513
• /
• 2006

본 논문에서는 출력평면에 나타나는 원 영상을 재생시키기 위하여 위상 랩핑 방법을 이용하여 암호화 수준을 향상시키고 주파수 영역에서 위상 부호화하여 암호화함으로써 잡음에 강한 복호화 방법을 제안하였다. 암호화된 영상은 원 영상이 아닌 위상 변조된 가상 영상과 무작위 위상 영상을 곱하고 제로 패딩(ze.o-padding)하여 푸리에 변환한 후 이 변환된 복소 영상을 데이터 전송 및 매핑을 용이하게 하기 위하여 실수 값으로 변환하여 위상 부호화하여 만든다. 복호화 과정은 제안한 선형적인 영상을 비선형적인 영상으로 변환시키는 위상 랩핑 방법에 의해 각각 만들어진 암호화된 영상과 복호화 키를 곱하여 푸리에 역변환하여 공간필터를 가진 출력 평면에서 원 영상을 복원함으로써 광축 정렬 문제와 픽셀 대 픽셀 대응이 용이하여 복원영상의 해상도를 향상시킬 수 있다. 제안한 방법은 허가되지 않은 사용자가 암호화된 영상을 분석함으로써 있을 수 있는 복제 가능성을 원 영상의 어떤 정보도 포함하지 않은 가상 영상을 사용함으로써 배제할 수 있고 또한 실수 값을 위상 부호화함으로써 현재에 사용되는 공간 광 변조기로 표현이 가능하다. 컴퓨터 모의 실험을 통하여 제안한 암호화 방법의 적합성과 암호화된 영상과 복호화 키 영상에 잡음이 발생하더라도 원 영상의 복원이 가능함을 확인하였다.

• Structural Engineering and Mechanics
• /
• 제17권6호
• /
• pp.819-828
• /
• 2004

A new process for estimating the natural frequency and the corresponding damping ratio in large structures is discussed. In a practical situation, it is very difficult to analyze large structures precisely because they are too complex to model using the finite element method and too heavy to excite using the exciting force method; in particular, the measured signals are seriously influenced by ambient noise. In order to identify the structural impulse response associated with the information of natural frequency and the corresponding damping ratio in large structures, the analysis process, a so-called "multiresolution blind system identification algorithm" which combines Mallat algorithm and the bicepstrum method. High time-frequency concentration is attained and the phase information is kept. The experimental result has demonstrated that the new analysis process exploiting the natural frequency and the corresponding damping ratio of structural response are useful tools in structural analysis application.

• 전력전자학회:학술대회논문집
• /
• 전력전자학회 1998년도 Proceedings ICPE 98 1998 International Conference on Power Electronics
• /
• pp.165-170
• /
• 1998

The paper describes developed method of feedforward digital modulation of line-to-line voltage of 3-phase inverter for drive application. It is based on representation of parameters of output voltage of inverter in function of operating frequency of drive system. Pure algebraic control laws and big computational simplicity characterize this scheme of modulation. It has been presented results of simulation of adjustable drive systems with the method of pulse-width modulation described.

• Journal of Power Electronics
• /
• 제15권5호
• /
• pp.1207-1216
• /
• 2015

In order to suppress the low frequency oscillation of the neutral-point voltage for three-level inverters, this paper proposes a new discontinuous pulse width modulation (DPWM) control method. The conventional sinusoidal pulse width modulation (SPWM) control has no effect on balancing the neutral-point voltage. Based on the basic control principle of DPWM, the relationship between the reference space voltage vector and the neutral-point current is analyzed. The proposed method suppresses the low frequency oscillation of the neutral-point voltage by keeping the switches of a certain phase no switching in one carrier cycle. So the operating time of the positive and negative small vectors is equal. Comparing with the conventional SPWM control method, the proposed DPWM control method suppresses the low frequency oscillation of the neutral-point voltage, decreases the output waveform harmonics, and increases both the output waveform quality and the system efficiency. An experiment has been realized by a neutral-point clamped (NPC) three-level inverter prototype based on STM32F407-CPLD. The experimental results verify the correctness of the theoretical analysis and the effectiveness of the proposed DPWM method.