• The Journal of the Acoustical Society of Korea
• /
• v.27 no.7
• /
• pp.386-393
• /
• 2008

This paper proposes a new approach for speech enhancement in highly nonstationary noisy environments. The spectral subtraction (SS) is a well known technique for speech enhancement in stationary noisy environments. However, in real world, noise is mostly nonstationary. The proposed method uses an auto control parameter for an adaptive threshold to work well in highly nonstationary noisy environments. Especially, the auto control parameter is affected by a linear function associated with an a posteriori signal to noise ratio (SNR) according to the increase or the decrease of the noise level. The proposed algorithm is combined with spectral subtraction (SS) using a hangover scheme (HO) for speech enhancement. The performances of the proposed method are evaluated ITU-T P.835 signal distortion (SIG) and the segment signal to-noise ratio (SNR) in various and highly nonstationary noisy environments and is superior to that of conventional spectral subtraction (SS) using a hangover (HO) and SS using a minimum statistics (MS) methods.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2008.04a
• /
• pp.742-746
• /
• 2008

With the trend of factory automation, nowadays, much industrial machinery tends to be put into 24-hours operation a day. However, these trends in industrial equipments also increase the possibility of various mechanical problems and bring about innumerable maintenance cost. There is a strong need of the condition monitoring and diagnosis for industrial equipment, especially rotating machinery, since they are connected not only to the reduction in the maintenance costs but also connected to the enhancement of production efficiency. Generally, to evaluate the operating conditions in the machinery in the industrial field, various physical properties are monitored. Among them, vibration and Noise signals are the mist important indicator and it is effectively used in many diagnosis systems for machinery. Much previous research is based in the FFT (Fast Fourier Transform) method. The spectral analysis is assumed that the signal is stationary. However, almost random signals are non-stationary. The wavelet transform has been recognized an efficient Method. Most interesting sounds have time-varying features. Signal processing techniques for the analysis of transient sound have been not clearly given yet.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37A no.12
• /
• pp.1122-1132
• /
• 2012

Compressed sensing has been applied to many fields such as images, speech signals, radars, etc. It has been mainly applied to stationary signals, and reconstruction error could grow as compression ratios are increased by decreasing measurements. To resolve the problem, speech signals are divided into frames and processed in parallel. The frames are made sparse by dictionary learning, and adaptive compressed sensing is applied which designs the compressed sensing reconstruction matrix adaptively by using the difference between the sparse coefficient vector and its reconstruction. Through the proposed method, we could see that fast and accurate reconstruction of non-stationary signals is possible with compressed sensing.

• Journal of Global Scholars of Marketing Science
• /
• v.16 no.1
• /
• pp.13-26
• /
• 2006

In this research, we develope a procedure for detecting a random non-stationarity to the individual's purchasing rate in a stationary NED model. On this purpose, we derive the likelihood ratio statistic for a testing null and alternative hypotheses defined as whether there is no significant structural change in a stationary NED model or any. Where the structural change comes from a random non-stationarity(marketing mix activities or seasonality, for example) to the individual's purchasing rate. We also apply the developed method to a panel data for a frequently purchased good. This research could be a solution to include the non-stationarity in a stationary NED model. We also expect that the developed model could give a signal for an early detection of significant changes in marketing environment, and a mean for a measurement of the effects of marketing mix activities.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.3E
• /
• pp.93-98
• /
• 2009

In this paper, we propose a new VSS-NLMS algorithm through a simple modification of the conventional NLMS algorithm, which leads to a low complexity algorithm with enhanced performance. The step size of the proposed algorithm becomes smaller as the error signal is getting orthogonal to the input vector. We also show that the proposed algorithm is an approximated normalized version of the KZ-algorithm and requires less computation than the KZ-algorithm. We carried out a performance comparison of the proposed algorithm with the conventional NLMS and other VSS algorithms using an adaptive channel equalization model. It is shown that the proposed algorithm presents good convergence characteristics under both stationary and non-stationary environments despites its low complexity.

• Smart Structures and Systems
• /
• v.2 no.1
• /
• pp.47-60
• /
• 2006

Bi-spectrum is a tool in the signal processing for identification of non-linear dynamic behvaiour in systems, and well-known for stationary system where components are non-linearly interacting. Breathing of a crack during shaft rotation is also exhibits a non-linear behaviour. The crack is known to generate 2X (twice the machine RPM) and higher harmonics in addition to 1X component in the shaft response during its rotation. Misaligned shaft also shows similar such feature as a crack in a shaft. The bi-spectrum method has now been applied on a small rotating rig to observe its features. The bi-spectrum results are found to be encouraging to distinguish these faults based on few experiments conducted on a small rig. The results are presented here.

• Journal of Biomedical Engineering Research
• /
• v.29 no.4
• /
• pp.278-285
• /
• 2008

In this study the complex-valued continuous wavelet transform (CWT) has been applied in detection of Electrocardiograms (ECG) as response to various signal classification methods such as Fourier transforms and other tools of time frequency analysis. Experiments have shown that CWT may serve as a detector of non-stationary signal changes as ECG. The tested signal is corrupted by short time events. We applied CWT to detect short-time event and the result image representation of the signal has showed us that one can easily find the discontinuity at the time scale representation. Analysis of ECG signal using complex-valued continuous wavelet transform is the first step to detect possible changes and alternans. In the second step, modulus and phase must be thoroughly examined. Thus, short time events in the ECG signal, and other important characteristic points such as frequency overlapping, wave onsets/offsets extrema and discontinuities even inflection points are found to be detectable. We have proved that the complex-valued CWT can be used as a powerful detector in ECG signal analysis.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.12
• /
• pp.1920-1925
• /
• 2012

In this paper, we propose a novel method to improve the performance of the improved minima controlled recursive averaging (IMCRA) in non-stationary noisy environment. The conventional IMCRA algorithm efficiently estimate the noise power by averaging past spectral power values based on a smoothing parameter that is adjusted by the signal presence probability in frequency subbands. Since the minimum of smoothing parameter is defined as 0.85, it is difficult to obtain the robust estimates of the noise power in non-stationary noisy environments that is rapidly changed the spectral characteristics such as babble noise. For this reason, we proposed the modified IMCRA, which adaptively estimate and updata the noise power according to the noise type classified by the Gaussian mixture model (GMM). The performances of the proposed method are evaluated by perceptual evaluation of speech quality (PESQ) and composite measure under various environments and better results compared with the conventional method are obtained.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.15 no.2 s.95
• /
• pp.192-198
• /
• 2005

Empirical mode decomposition(EMD) method has been recently proposed to analyze non-linear and non-stationary data. This method allows the decomposition of one-dimensional signals into intrinsic mode functions(IMFs) and is used to calculate a meaningful multi-component instantaneous frequency. In this paper, it is assumed that each mode of damped vibration signal could be well separated in the form of IMF by EMD. In this case, we can have a new powerful method to calculate natural frequencies and dampings from damped vibration signal which usually has multiple modes. This proposed method has been verified by both simulation and experiment. The results by EMD method whichhas used only output vibration data are almost identical to the results by FRF method which has used both input and output data, thereby proving usefulness and accuracy of the proposed method.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2004.11a
• /
• pp.699-704
• /
• 2004

Empirical mode decomposition(EMD) method has been recently proposed to analyze non-linear and non-stationary data. This method allows the decomposition of one-dimensional signals into intrinsic mode functions(IMFs) and is used to calculate a meaningful multi-component instantaneous frequency. In this paper, it is assumed that each mode of damped vibration signal could be well separated in the form of IMF by EMD. In this case, we can have a new powerful method to calculate natural frequencies and dampings from damped vibration signal which usually has multiple modes. This proposed method has been verified by both simulation and experiment. The result by EMD method which has used only output vibration data is almost identical to the result by FRF method which has used both input and output data, thereby proving usefulness and accuracy of the proposed method.