• 정보처리학회논문지A
• /
• 제18A권1호
• /
• pp.1-10
• /
• 2011

물리적 모델링은 실제 악기음과 유사한 고음질의 음을 합성하는 방법으로 많은 연구가 진행되어 왔다. 그러나 물리적 모델링은 악기의 소리를 합성할 때 필요한 수많은 파라미터들을 동시에 계산해야 하기 때문에 동시 발음수가 높은 악기의 경우 실시간 처리에 문제가 발생할 수 있다. 이러한 문제를 해결하기 위해 본 논문에서는 전통 현악기인 가야금의 음 합성 알고리즘을 실시간으로 처리 가능한 단일 명령어 다중 데이터(Single Instruction Multiple Data, SIMD) 방식의 멀티코어 프로세서를 제안한다. 제안하는 SIMD기반 멀티코어 프로세서는 가야금의 12개현을 제어하기 위해 12개의 프로세싱 엘리먼트(Processing Element, PE)로 구성되어 있다. 각각의 프로세싱 엘리먼트는 해당되는 가야금 현을 모델링하며, 각 현의 여기신호와 파라미터를 음 합성 병렬 알고리즘의 입력으로 받아 동시에 12개 현의 합성된 음을 실시간으로 생성할 수 있다. 표본화 비율을 44.1kHz로 설정하고 16비트 양자화 데이터의 음을 합성한 모의실험 결과, 제안한 SIMD기반 멀티코어 프로세서를 이용한 합성음은 원음과 매우 유사하였으며, 상용 프로세서(TI TMS320C6416, ARM926EJ-S, ARM1020E)보다 실행 시간에서 5.6~11.4배, 에너지 효율에서 553~1,424배의 향상을 보였다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2002년도 추계학술대회논문집
• /
• pp.945-949
• /
• 2002

This part of specifies the method for determining the dynamic stiffness of resilient materials used under floating floors. Dynamic stiffness is one of the parameters that determine the sound insulation of such floors in dwellings. This part applies to the determination of dynamic stiffness per unit area of resilient materials with smooth surfaces used in a continuous layer under floating floors in dwellings

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2002년도 추계학술대회논문초록집
• /
• pp.392.1-392
• /
• 2002

This part of specifies the method for determining the dynamic stiffness of resilient materials used under floating floors. Dynamic stillness is one of the parameters that determine the sound insulation of such floors in dwellings. This part applies to the determination of dynamic stiffness per unit area of resilient materials with smooth surfaces used in a continuous layer under floating floors in dwellings.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2005년도 추계종합학술대회
• /
• pp.1161-1164
• /
• 2005

In this paper, we proposed the technique to estimate the life cycle of Internet content services based on the logistic regression model. In this paper, to define parameters of Internet contents estimating life cycle by logistic regression model, we used market size, traffic amount, page view and session-visit number as the parameters of Internet contents estimating life cycle by logistic regression model. In this paper, to compare the performance of our proposed scheme, we estimated life cycle for the download services of bell sound & character contents in mobile network. As a result, using our proposed logistic regression, we were able to estimate exactly the life cycle of the download services of bell sound & character contents.

• Journal of Biosystems Engineering
• /
• 제23권1호
• /
• pp.57-66
• /
• 1998

This study was conducted to investigate the relationships between parameters affecting internal quality of watermelon and its acoustic characteristics. The measuring and analyzing system was established to study the acoustic characteristics of watermelon. Algorithms for analyzing sound signals were developed. Sound signals which was detected with the microphone were filtered, and their spectrum was computed by means of the Fast Fourier Transform. As watermelon changed from the unripe stage to ripe, acoustic waves in time domain became complicated, and several components appeared in frequency domain. The correlationship was investigated between some parameters affecting internal quality of the watermelon and several peak frequencies. Results indicated that weight, density and sugar content had high correlations with several frequencies(the first peak frequency, the second peak frequency, and the third peak frequency). And the sugar content and the volume of the watermelon were highly correlated with the third peak frequency.

• 한국컴퓨터산업학회논문지
• /
• 제2권10호
• /
• pp.1245-1252
• /
• 2001

본 연구에서는 한국어의 숫자음을 대상 어휘로 이용하여 화자 식별을 하였다. 화자 식별을 위해서는 음성파에 포함된 여러 가지 정보로부터 화자 개개인이 갖는 개인적 특정 정보를 추출하여 화자 식별을 위한 특징 파라메타로 이용한다. 실험에서 사용한 특징 파라메타로는 음성파의 피치주기 사이에 존재하는 멀티펄스가 화자에 의존하는 개인의 특정 정보를 포함하고 있음에 착안하여 개개인의 음성파로부터 멀티펄스를 추출하고 추출된 멀티펄스로부터 화자별 특징 파라메타를 탐색하여 화자 식별 실험을 하였다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2002년도 추계학술대회논문초록집
• /
• pp.333.1-333
• /
• 2002

This paper suggests the noise reduction method of the engine cooling fan. It was estimated the fannoise contribution at the engine room and identified the noise source at the rotating fan by sound intensity method, first. And it has been developed the program for predicting the noise spectrum of axial flow fan. The radiated acoustic pressure is expressed the discrete frequency noise peaks at BPF and its harmonics and the line spectrum at the broad band by the noise generation mechanisms. In this paper it is shown that the comparison of the measuted and calculaed noise spectra of fn for the validation of the noise predictiong program. And this paper presents the characteristics of a fan noise due to modify the design parameters. Accordingly, it was obtained the design parameter values for noise reduction of fan.

• The Journal of the Acoustical Society of Korea
• /
• 제29권2E호
• /
• pp.64-72
• /
• 2010

Osteoporosis is a skeletal disease characterized by two factors: reduced bone mass and microstructure disruption of bone tissue. These symptoms increase bone fragility and can contribute to eventual fracture. In recent years, quantitative ultrasound (QUS) technologies have played a growing role in the diagnosis of osteoporosis. Most of the commercial bone sonometers measure speed of sound and/or broadband ultrasound attenuation at peripheral skeletal sites. However, QUS parameters are purely empirical measures that have not yet been firmly linked to physical parameters, such as bone strength or porosity, and the underlying physics for their variations in cancellous bone is not well understood yet. This paper reviews the QUS technologies for the diagnosis of osteoporosis and also addresses several theoretical models, such as the Biot model, the scattering model, the stratified model, and the modified Biot-Attenborough model, for ultrasonic wave propagation in bone.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2004년도 추계학술대회논문집
• /
• pp.317-322
• /
• 2004

In this paper, It was describe the processes of development of HiFi speaker system. Woofer and tweeter were fabricated by unskilled students and their 1.5 parameters were identified by known mass method. Based on T-S parameters port enclosure was designed and built by means of software. Acoustic radiation phenomena of port enclosure were simulated and compared to test result. Acoustic pressure difference between lower frequency and higher frequency was flattened by adopting optimal crossover network. Finally, built HiFi speaker system was showed good sound quality and sound pressure and electrical impedance was well agreed with test results each other.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1994년도 FIFTH WESTERN PACIFIC REGIONAL ACOUSTICS CONFERENCE SEOUL KOREA
• /
• pp.686-691
• /
• 1994

For evaluating the response fluctuation of the actual environmental acoustic system excited by arbitrary random inputs, it is important to predict a whole probability distribution form closely connected with evaluation indexes Lx, Leq and so on. In this paper, a new type evaluation method is proposed by introducing three functional models matched to the prediction of the response probability distribution from a problem-oriented viewpoint. Because of the positive variable of the sound intensity, the response probability density function can be reasonably expressed theoretically by a statistical Laguerre expansion series form. The relationship between input and output is described by the regression relationship between the distribution parameters(containing expansion coefficients of this expression) and the stochastic input. These regression functions are expressed in terms of the orthogonal series expansion and their parameters are determined based on the least-squares error criterion and the measure of statistical independency.