• The Journal of the Acoustical Society of Korea
• /
• 제22권2E호
• /
• pp.87-92
• /
• 2003

In this paper, a new speech enhancement method using level adapted wavelet packet is presented. First, we propose a level adapted wavelet packet to alleviate a drawback of the conventional node adapted one in noisy environment. Next, we suggest an adaptive noise estimation method at each node on level adapted wavelet packet tree. Then, for more accurate noise component subtraction, we propose a new estimation method of spectral subtraction weight. Finally, we present a modified spectral subtraction method. The proposed method is evaluated on various noise conditions: speech babble noise, F-l6 cockpit noise, factory noise, pink noise, and Volvo car interior noise. For an objective evaluation, the SNR test was performed. Also, spectrogram test and a very simple listening test as a subjective evaluation were performed.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2008년도 춘계학술대회논문집
• /
• pp.496-499
• /
• 2008

The noise reducing devices installed on the noise barrier have been developed in many shapes and ways to reduce noise around road traffic areas. In this study, test and estimation method for the noise reducing device witch installed on the top of a noise barrier was suggested. For this, the authors have considered sound power flow around the device and sound pressure levels for the far field area. To estimate the area effect behind the barrier, area average of noise pressure level difference divided by two area, upper and bellow the sight-line. Comparing the attenuation difference of these areas, the tendency of noise reduction effect was studied according to type of noise reducing devices. Compared with noise shielding efficiency of the devices that using equivalent height of a simple barrier calculated by the SoundPlan, the commercial environment noise simulation software.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2006년도 춘계학술대회논문집
• /
• pp.1129-1134
• /
• 2006

Mass estimation was derived as functions of acceleration magnitude and primary frequency. The conventional method of mass estimation used frequency data directly in the frequency domain. The signals that can be obtained sensor contained noise as well as impact signal. Therefore, how well we can detect the frequency data in noise directly determines the quality of mass estimation. To find exact frequency data, we used time-frequency analysis. The time frequency method are expected to be more useful than the conventional frequency domain analyses for the mass estimation problem on a plate type structure. Also it has been concluded that the smoothed WVD can give more reliable means than the other methodologies for the mass estimation in a noisy environment.

• 한국소음진동공학회논문집
• /
• 제16권8호
• /
• pp.872-878
• /
• 2006

Mass estimation was derived as functions of acceleration magnitude and primary frequency. The conventional method of mass estimation used frequency data directly in the frequency domain. The signals that can be obtained sensor contained noise as well as impact signal. Therefore, how well we can detect the frequency data in noise directly determines the quality of mass estimation. To find exact frequency data, we used time-frequency analysis. The time-frequency methods are expected to be more useful than the conventional frequency domain analyses for the mass estimation problem on a plate type structure. Also it has been concluded that the smoothed WVD can give more reliable means than the other methodologies for the mass estimation in a noisy environment.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2011년도 춘계학술대회 논문집
• /
• pp.580-584
• /
• 2011

Sound power is one of the most significant factors to predict and assess noise from sound source. However, many researchers developed indirect methods to calculate the sound power with noise level because it is impossible to measure sound power directly while a train is running. In this paper, we proposes a method to estimate separately each sound power for rolling noise and propulsive noise with the noise emitted from rolling stocks and verifies the validity of the estimation method using other sound power estimation formula.

• 한국정보통신학회논문지
• /
• 제26권12호
• /
• pp.1926-1933
• /
• 2022

변수내장형 다채널 위너 필터는 내장된 변수를 이용하여 잔여잡음과 신호왜곡 간의 트레이드오프를 조절할 수 있는 선형 필터이다. 이러한 변수내장형 다채널 위너필터를 적용하기 위해서는 정확한 잡음추정이 중요한데 널리 쓰이는 다채널 최소 제어 재귀 평균 기법이 있다. 하지만 다채널 최소 제어 재귀 평균 기법은 방향성 간섭 신호가 존재할 경우 잡음추정의 정확도가 하락하여 변수내장형 다채널 위너필터의 성능이 저하되는 문제점이 있다. 따라서, 본 논문에서는 변수내장형 다채널 위너필터를 위한 새로운 잡음 추정 기법을 제안한다. 제안한 방법은 주로 잡음 섞인 마이크로폰 입력 신호의 전력 스펙트럼 밀도에 대해 고유값 분해, 방향성 정보를 이용한 목적신호의 기여도 추정, 목적신호의 기여도를 보다 합리적으로 추정하기 위한 지수 가중치 부가의 일련의 과정을 수행한다. 제안한 방법을 평가하기 위해 신호대잡음비, 음성왜곡도 등의 총 4가지 객관적 성능 평가 방법을 이용하여 기존의 방법과 비교하였다. 실험을 통해 방향성 간섭신호가 존재하는 환경에서 제안한 잡음 추정기법을 적용한 다채널 위너필터의 성능이 향상됨을 확인하였다.

• 전자공학회논문지B
• /
• 제32B권1호
• /
• pp.85-95
• /
• 1995

The error intensity models for the ML estimation of a signal parameter have been developed in a companion paper [1]. While the methods described in [1] are applicable to any estimation problem with continuous parameters, our main application in this paper is the time delay estimation, and comparisons among the models derived in [1] (i.e. LC, LM, and ALM models)have been made. We first consider the case where only additive Gaussian noise is involved, and then the shot noise environment where coherent impulsive noise is also involved in addition to the Gaussian noise. We compare the models in terms of the probability of error, MSE(Mean Squared Error), and the computational complexity, which are the most important performance criteria in the analysis of parameter estimation. In conclusion, the ALM model turned out to be the most adequate model of all from the viewpoints of the criteria mentioned above.

• 대한전자공학회논문지SP
• /
• 제42권3호
• /
• pp.129-136
• /
• 2005

잡음이 중첩된 음성으로부터 잡음을 제거하기 위해서는, 잡음의 크기에 따라서 음성처리 시스템의 매개변수를 변경하는 것이 양호한 음질의 음성을 재생하는데 바람직하다. 본 논문은 백색잡음 및 자동차의 주행잡음에 의해 저하된 3단계의 음성을 학습할 수 있는 3층 구조의 신경회로망을 사용하여, 음성 중의 잡음량의 크기를 추정하는 방식을 제안한다. 실험결과, 제안한 방법은 신경회로망에 의해서 잡음량이 추정될 수 있는 것을 알 수 있었으며, 화자와 음성 데이터가 학습데이터와 다르더라도 백색잡음에 대해서 평균 $95\%$ 이상의 높은 잡음 추정율을 구할 수 있었다.

• IEIE Transactions on Smart Processing and Computing
• /
• 제3권2호
• /
• pp.41-51
• /
• 2014

This paper describes a method to estimate the noise power using the minimum statistics approach, which was originally proposed for audio processing. The proposed minimum statistics-based method separates a noisy image into multiple frequency bands using the three-level discrete wavelet transform. By assuming that the output of the high-pass filter contains both signal detail and noise, the proposed algorithm extracts the region of pure noise from the high frequency band using an appropriate threshold. The region of pure noise, which is free from the signal detail part and the DC component, is well suited for minimum statistics condition, where the noise power can be extracted easily. The proposed algorithm reduces the computational load significantly through the use of a simple processing architecture without iteration with an estimation accuracy greater than 90% for strong noise at 0 to 40dB SNR of the input image. Furthermore, the well restored image can be obtained using the estimated noise power information in parametric image restoration algorithms, such as the classical parametric Wiener or ForWaRD image restoration filters. The experimental results show that the proposed algorithm can estimate the noise power accurately, and is particularly suitable for fast, low-cost image restoration or enhancement applications.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.612-617
• /
• 2009

In a DCT coding, degradations called block artifact and mosquito noise are appeared in reconstructed pictures. They should be reduced in post processing after decoding without superabundant processing. However, an estimation of mosquito noise is rare because of its difficulty. To realize an estimation of mosquito noise level, we extract a block that mosquito noise will be easy to occur. Mosquito noise level is calculated at a selected side of the block. In this processing, only the sides of high probability block are used. Then, a block value is taken by averaging. Finally, the picture value is calculated by averaging of this. Estimation method is evaluated by using the MPEG-4 decoded pictures. Quantization scale of coding and the estimated mosquito noise level are compared. As the results, we recognize the proposed method gives almost reasonable mosquito block and absolute level. Father, adaptive filter is controlled by the estimated mosquito noise level. It is recognized that the high quality of decoded picture is kept and the mosquito noise is reduced effectively at the picture with degradation.