• 지구물리와물리탐사
• /
• 제10권2호
• /
• pp.131-137
• /
• 2007

초동발췌 방법과 이와 관련된 오차문제를 고찰하고 자료처리 전산화 측면에서 초동발췌 및 오차계산의 실용적 알고리즘을 제안하였다. 제안한 초동발췌 방법은 2단계로 구성된다. 1) 초동신호 중 트레이스 간 연속성이 양호한 최초의 피크(peak) 또는 트라프(trough)를 발췌한다. 2) 발췌시점 전방 일정구간의 기록을 직선으로 근사화 하고 이 직선의 시간 절편을 초동주시로 한다. 근사화 구간의 길이는 대략 초동 웨이브렛 폭의 1/4보다 다소 작게 설정한다. 초동발췌에 내재된 오차의 정량적인 척도는 기록의 특정시점에 초동의 도착 여부를 판단하는 데 필요한 기록의 길이로 정의한다. 기록의 길이를 나타내는 공식은 신호의 주파수대역과 신호대잡음비의 함수로 표시된다. 3개의 공대공 탄성파기록의 초동을 수동 및 제안한 방법으로 발췌하였고 각 트레이스의 오차한계를 계산하였다. 실험결과 제안한 근사직선 시간절편법의 우수한 성능과 발췌된 초동평가에 있어서 발췌오차의 유용성을 확인하였다.

• 말소리와 음성과학
• /
• 제4권4호
• /
• pp.79-86
• /
• 2012

In the two-band excitation (TBE) model, maximum voiced frequency (MVF) is the most important feature of the excitation parameter because the synthetic speech quality depends on MVF. Thus, this paper proposes an enhanced MVF estimation scheme based on the peak picking method. In the proposed scheme, the local peak and the peak lobe are picked from the spectrum of a linear predictive residual signal. The normalized distance between neighboring peak lobes is calculated and utilized as a feature to estimate MVF. Experimental results of both objective and subjective tests show that the proposed scheme improves synthetic speech quality compared with that of the conventional one.

• Smart Structures and Systems
• /
• 제31권1호
• /
• pp.45-56
• /
• 2023

Vibration-based structural health monitoring is used to ensure the safety of structures by installing sensors in structures. The peak picking method, one of the applications of vibration-based structural health monitoring, is a method that analyze the dynamic characteristics of a structure using the peaks of the frequency response function. However, the results may vary depending on the person predicting the peak point; further, the method does not predict the exact peak point in the presence of noise. To overcome the limitations of the existing peak picking methods, this study proposes a new method to automate the modal analysis process by utilizing long short-term memory, a type of recurrent neural network. The method proposed in this study uses the time series data of the frequency response function directly as the input of the LSTM network. In addition, the proposed method improved the accuracy by using the phase as well as amplitude information of the frequency response function. Simulation experiments and lab-scale model experiments are performed to verify the performance of the LSTM network developed in this study. The result reported a modal assurance criterion of 0.8107, and it is expected that the dynamic characteristics of a civil structure can be predicted with high accuracy using data without experts.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제51권2호
• /
• pp.86-88
• /
• 2002

An estimation method for the characteristic parameters in the low frequency signal is proposed in this paper. A low frequency signal is assumed to be modulated or distorted by high frequency terms. The algorithm proposed in this paper is designed to select set of local maximums in a successive manner, hence it is denoted as the iterative peak picking(IPP) algorithm. The IPP algorithm is operating in the time domain and is using only the comparison operation between two neighboring samples. Therefore, its computational complexity is very low and the delay caused by the computation is negligible, which make the real-time operation possible with economic hardware. The proposed algorithm is verified on the pitch estimation of speech signal and blood pulse estimation.

• Structural Engineering and Mechanics
• /
• 제17권3_4호
• /
• pp.429-444
• /
• 2004

The ambient vibration measurement is a kind of output data-only dynamic testing where the traffics and winds are used as agents responsible for natural or environmental excitation. Therefore an experimental modal analysis procedure for ambient vibration testing will need to base itself on output-only data. The modal analysis involving output-only measurements presents a challenge that requires the use of special modal identification technique, which can deal with very small magnitude of ambient vibration contaminated by noise. Two complementary modal analysis methods are implemented. They are rather simple peak picking (PP) method in frequency domain and more advanced stochastic subspace identification (SSI) method in time domain. This paper presents the application of ambient vibration testing and experimental modal analysis on large civil engineering structures. A 15 storey reinforced concrete shear core building and a concrete filled steel tubular arch bridge have been chosen as two case studies. The results have shown that both techniques can identify the frequencies effectively. The stochastic subspace identification technique can detect frequencies that may possibly be missed by the peak picking method and gives a more reasonable mode shapes in most cases.

• ETRI Journal
• /
• 제37권6호
• /
• pp.1211-1219
• /
• 2015

In a hidden Markov model-based speech synthesis system using a two-band excitation model, a maximum voiced frequency (MVF) is the most important feature as an excitation parameter because the synthetic speech quality depends on the MVF. This paper proposes an enhanced MVF estimation scheme based on a peak picking method. In the proposed scheme, both local peaks and peak lobes are picked from the spectrum of a linear predictive residual signal. The average of the normalized distances of local peaks and peak lobes is calculated and utilized as a feature to estimate an MVF. Experimental results of both objective and subjective tests show that the proposed scheme improves the synthetic speech quality compared with that of a conventional one in a mobile device as well as a PC environment.

• 한국음향학회지
• /
• 제24권5호
• /
• pp.248-254
• /
• 2005

본 논문에서는 대역분할 광대역 음성 부호화기의 구조와 음질 향상을 위한 새로운 고대역 구조를 제안한다. 대역분할 방식에 의해 광대역 음성은 저대역 ($O\~4kHz$) 음성과 고대역 ($4\~8kHz$) 음성으로 나뉘어 지고 각각 G.729E와 MLT(Modulated Lapped Transform) 여기모델을 적용하여 서로 독립된 방식으로 부호화한다. 4kbps의 낮은 전송률로 부호화되는 고대역에서는 MLT 여기모델을 효율적으로 이용하기 위하여 유 무성음을 구별하였고 유성음에 대해서는 저대역 피치주기를 이용한 MLT peak picking 방법을 적용하였다. 즉, MLT 변환된 여기신호는 주기적인 피크를 갖는 주기신호로 나타나며 이때의 피크값을 추출하여 양자화하여 전송한다. 무성음에 대해서는 에너지 값에 따라 비트를 달리 적용하고, 선형예측 스펙트럴 응답이 가중된 MLT 벡터 양자화 방법을 적용하였다. 제안된 15.8kbps 광대역 음성 부호화기의 성능평가는 주관적인 음질평가로 선호도 테스트를 수행하였다.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2001년도 봄 학술발표회 논문집
• /
• pp.439-445
• /
• 2001

In this paper, several modal identification techniques without using the input information are investigated. Generally, the peak picking method is most widely used, however, other methods may give better estimates particularly for the cases with close modes and/ or highly damped system modes. Example analyses were carried out on three different structures, and the estimated modal parameters by various methods are compared.

• Smart Structures and Systems
• /
• 제6권4호
• /
• pp.373-388
• /
• 2010

The updated finite element model of K$\ddot{o}$m$\ddot{u}$rhan Highway Bridge on the Firat River located on the $51^{st}$ km of Elazi$\breve{g}$-Malatya highway is obtained by using analytical and experimental results. The 2D and 3D finite element model of the bridge is created by using SAP2000 structural analyses software, and the dynamic characteristics of the bridge are determined analytically. The experimental measurements are carried out by Operational Modal Analysis Method under traffic induced vibrations and the dynamic characteristics are obtained experimentally. The vibration data are gathered from the both box girder and the deck of the bridge, separately. Due to the expansion joint in the middle of the bridge, special measurement points are selected when experimental test setups constitute. Measurement duration, frequency span and effective mode number are determined by considering similar studies in literature. The Peak Picking method in the frequency domain is used in the modal identification. At the end of the study, analytical and experimental dynamic characteristic are compared with each other and the finite element model of the bridge is updated by changing some uncertain parameters such as material properties and boundary conditions. Maximum differences between the natural frequencies are reduced from 10% to 2%, and a good agreement is found between natural frequencies and mode shapes after model updating.

• 전기학회논문지
• /
• 제64권11호
• /
• pp.1605-1610
• /
• 2015

This study describes a new method to analyze phonocardiogram acquired from electronic stethoscope. The method uses the formant frequencies of linear prediction spectrum of the phonocardiogram and proposes a novel method for formant detection using the smoothing and the first and second derivatives. For this, stethoscope sounds are acquired in university hospital. The stethoscope signals are preprocessed and analyzed by the Burg algorithm, a kind of linear prediction analysis. Based on the linear prediction spectra, the formant frequencies are estimated. The proposed method has shown better performance in formant frequency detection than the conventional peak picking method.