• The Journal of the Acoustical Society of Korea
• /
• v.38 no.3
• /
• pp.321-327
• /
• 2019

The purpose of this study is paper is to improve the flow performance and to reduce the aerodynamic noise of air discharge system consisting of a centrifugal fan, ducts and a housing for the clothes dryer. Using computational fluid dynamics and acoustic analogy based on FW-H (Ffowcs-Williams and Hawkings) Eq., air flow field and acoustic fields of the air discharge system are investigated. To optimize aerodynamic performance and aerodynamic noise, the response surface method is employed. The two factors central composite design using the inflow and outflow angles of fan blades is adopted. The devised optimum design shows the reduction of turbulent kinetic energy in the ducts and the housing of the system, and as a result, the improved flow rate and reduce noise is confirmed. Finally, the experment using the proto-type manufactured usign the optimum design shows the increase of flow rate by 4.2 %.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.8
• /
• pp.375-382
• /
• 2006

This paper is about how to identify emotional information and how to detect a specific emotion from speech signals. For emotion identification and detection task. we use long-term acoustic feature parameters and select the optimal Parameters using the feature selection technique based on F-score. We transform the conventional SVM into probabilistic output SVM for our emotion identification and detection system. In this paper we propose three approximation methods for log-likelihoods in a hypothesis test and compare the performance of those three methods. Experimental results using the SUSAS database showed the effectiveness of both feature selection and Probabilistic output SVM in the emotion identification task. The proposed methods could detect anger emotion with 91.3% correctness.

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.6
• /
• pp.404-411
• /
• 2018

This paper proposes a dolphin whistle mimicking underwater communication method using FSK (Frequency Shift Keying) and method to improve BER (Bit Error Rate) performance by using multipath gain combining. The proposed method divides whistle sound into short time intervals and transmits FSK modulated signal that ensures orthogonality of the symbol. Multipath gain can be obtained by using characteristic of mimicked signal frequency that varies with time. To demonstrate the performance of the proposed method, computer simulations and lake experiments were conducted. Computer simulation results show that an additional multipath gain is obtained by multipath. From lake experiments, when symbol length is 20 msec and modulation band is 900 Hz, the proposed FSK method with multipath combining gain obtains BER of 0.002, which is better than CSS (Chirp Spread Spectrum) with BER of 0.185. he proposed based on FSK method has higher imitation degree than the CSS method by analyzing mean cross-correlation value in the time - frequency domain of the imitated signal and actual whistle signal.

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.2
• /
• pp.83-91
• /
• 2018

The noise sources of a tip-jet driven rotor can be separated by rotor blade noise and jet noise. The rotor blade noise consists of thickness noise, loading noise, nonlinear quadrupole noise, and jet noise is divided into nozzle momentum noise and jet radiation noise. The flow analysis for the prediction of rotor blade noise is performed by CFD (Computational Fluid Dynamics) analysis, and the noise source of the rotor blade noise is identified by simultaneously applying the permeable and impermeable surface based FW-H (Ffowcs Williams-Hawkings) acoustic analogy. The nozzle momentum noise is obtained by permeable surface FW-H, and jet radiation noise is predicted by using empirical method for the fixed-wing jet. Both of jet noises use nozzle exit condition for noise analysis. The accuracy of the technique is verified based on the noise measurements of the tip-jet driven rotor, and the unique noise characteristics of the tip-jet driven rotor is confirmed by spectrum analysis.

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.6
• /
• pp.396-403
• /
• 2018

In this paper, a high-performance and low-noise centrifugal fan is developed for cooling automotive seats which provide a driver with pleasant driving environment. First, the flow characteristics of the existing fan unit was analyzed using a fan performance tester and CFD (Computational Fluid Dynamics) simulations. The analysis of the predicted flow field indicated vortex flow near the tip of fan hub and stagnation flow on the top of fan hub. Two design points are devised to reduce the vortex flow and the stagnation flow observed in the existing fan unit. First, the cut-off clearance which is the minimum distance between the fan blade and the fan housing is increased to reduce the vortex strength and, as a result, to reduce the overall sound pressure level. Second, the hub shape is more modified to eliminate the stagnation flow. The validity of proposed design is confirmed through the numerical analysis. Finally, a prototype is manufactured with a basis on the numerical analysis result and its improved flow and noise performances are confirmed through the P-Q curves measured by using the Fan Tester and the SPL (Sound Pressure Level) levels measured in the anechoic chamber.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.3
• /
• pp.213-221
• /
• 2009

The time reversal mirror (TRM) method for underwater communications has been developed to improve transmission performance with low complexity. However, digital communication parameters for TRM have not been researched deeply. This paper demonstrates that the TRM scheme obtains spatial diversity gain similar to multiple antennas, and proposes design methodologies of symbol interval, frame duration and transmission protocol for time reversal mirror transmission. Simulation results show that spatial diversity gain is achieved and the effect of ISI decreases as the number of transducer increases.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.5
• /
• pp.580-586
• /
• 2019

In this paper, we propose a biomimetic communication method using a dolphin whistle to covertly transmit the communication signal. A conventional CSS (Chirp Spread Spectrum) modulation technique divides dolphin whistle into several slots and modulates with up and down chirp signals. That causes the time-frequency characteristic difference between the original dolphin whistle and the camouflage performance is degraded. In this paper, we propose a delay based modulation scheme to eliminate distortions. The simulation results show that the bit error rate of the proposed method is better performance than that of the conventional CSS modulation method by about 3.5 dB to 8 dB. And the camouflage performance that evaluated through the cross correlation in the time-frequency domain is also better than that of the CSS modulation method.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.3
• /
• pp.219-227
• /
• 2021

This paper proposed an Orthogonal Frequency Division Multiplexing (OFDM) based biomimetic communication method using a dolphin whistle which covertly transmits communication signals to allies. The proposed method divides the dolphin whistle into several time slots corresponding to a number of OFDM symbols, and modulates the communication signal by mapping differential phase shift keying (DPSK) symbols into subcarriers that have the frequency bands of the dolphin whistle in each slot. The advantages of the proposed method are as follows: In the conventional Chirp Spread Spectrum (CSS) and Frequency Shift Keying (FSK) based biomimetic communication methods, the discontinuity of the frequency contour is large, but the proposed method can reduce the discontinuity. Even if the modulation order is increased, the degradation of the mimicking performance is small. The computer simulations demonstrate that the Bit Error Rate (BER) and mimicking performance of the proposed method are better performance than those of the conventional CSS and FSK.

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.2
• /
• pp.92-98
• /
• 2018

This paper is a study on the generation mechanism of propeller singing based on the cavitation tunnel test, underwater impact test, finite element analysis and computational flow analysis for the model propeller. A wire screen mesh, a propeller and a rudder were installed to simulate ship stern flow, and occurrence and disappearance of propeller singing phenomenon were measured by hydrophone and accelerometer. The natural frequencies of propeller blades were predicted through finite element analysis and verified by contact and non-contact impact tests. The flow velocity and effective angle of attack for each section of the propeller blades were calculated using RANS (Reynolds Averaged Navier-Stokes) equation-based computational fluid analysis. Using the high resolution analysis based on detached eddy simulation, the vortex shedding frequency calculation was performed. The numerical predicted vortex shedding frequency was confirmed to be consistent with the singing frequency and blade natural frequency measured by the model test.

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.6
• /
• pp.475-482
• /
• 2018

In this paper, we present a study on data augmentation methods for DNN (Deep Neural Network)-based audio tagging. In this system, an audio signal is converted into a mel-spectrogram and used as an input to the DNN for audio tagging. To cope with the problem associated with a small number of training data, we augment the training samples using time stretching, pitch shifting, dynamic range compression, and block mixing. In this paper, we derive optimal parameters and combinations for the augmentation methods through audio tagging simulations.