• 한국음향학회:학술대회논문집
• /
• 한국음향학회 2002년도 하계학술발표대회 논문집 제21권 1호
• /
• pp.9-14
• /
• 2002

We propose a new algorithm for blind source separation (BSS), in which frequency-domain independent component analysis (FDICA) and time-domain ICA (TDICA) are combined to achieve a superior source-separation performance under reverberant conditions. Generally speaking, conventional TDICA fails to separate source signals under heavily reverberant conditions because of the low convergence in the iterative learning of the inverse of the mixing system. On the other hand, the separation performance of conventional FDICA also degrades significantly because the independence assumption of narrow-band signals collapses when the number of subbands increases. In the proposed method, the separated signals of FDICA are regarded as the input signals for TDICA, and we can remove the residual crosstalk components of FDICA by using TDICA. The experimental results obtained under the reverberant condition reveal that the separation performance of the proposed method is superior to that of conventional ICA-based BSS methods.

• The Journal of the Acoustical Society of Korea
• /
• 제23권2E호
• /
• pp.56-67
• /
• 2004

Despite abundant research outcomes of blind source separation (BSS) in many types of simulated environments, their performances are still not satisfactory to be applied to the real environments. The major obstacle may seem the finite filter length of the assumed mixing model and the nonlinear sensor noises. This paper presents a two-step speech enhancement method with multiple microphone inputs. The first step performs a frequency-domain BSS algorithm to produce multiple outputs without any prior knowledge of the mixed source signals. The second step further removes the remaining cross-channel interference by a spectral cancellation approach using a probabilistic source absence/presence detection technique. The desired primary source is detected every frame of the signal, and the secondary source is estimated in the power spectral domain using the other BSS output as a reference interfering source. Then the estimated secondary source is subtracted to reduce the cross-channel interference. Our experimental results show good separation enhancement performances on the real recordings of speech and music signals compared to the conventional BSS methods.

• 전기학회논문지
• /
• 제57권8호
• /
• pp.1454-1463
• /
• 2008

Blind source separation by independent component analysis (ICA) has applied in signal processing, telecommunication, and image processing to recover unknown original source signals from mutually independent observation signals. Neural networks are learned to estimate the original signals by unsupervised learning algorithm. Because the outputs of the neural networks which yield original source signals are mutually independent, then mutual information is zero. This is equivalent to minimizing the Kullback-Leibler convergence between probability density function and the corresponding factorial distribution of the output in neural networks. In this paper, we present a learning algorithm using information theory and higher order statistics to solve problem of blind source separation. For computer simulation two deterministic signals and a Gaussian noise are used as original source signals. We also test the proposed algorithm by applying it to several discrete images.

• Smart Structures and Systems
• /
• 제26권2호
• /
• pp.135-146
• /
• 2020

This paper deals with an application of adaptive blind source separation (BSS) method, equivariant adaptive separation via independence (EASI), and Teager Energy Operator (TEO) for online identification of structural modal parameters. The aim of adaptive BSS methods is recovering a set of independent sources from their unknown linear mixtures in each step when a new sample is received. In the proposed approach, firstly, the EASI method is used to decompose structural responses into independent sources at each instance. Secondly, the TEO based demodulation method with discrete energy separation algorithm (DESA-1) is applied to each independent source, and the instantaneous frequencies and damping ratios are extracted. The DESA-1 method can provide the fast time response and has high resolution so it is suitable for online problems. This paper also compares the performance of DESA-1 algorithm with Hilbert transform (HT) method. Compared to HT method, the DESA-1 method requires smaller amounts of samples to estimate and has a smaller computational complexity and faster adaption due to instantaneous characteristic. Furthermore, due to high resolution of the DESA-1 algorithm, it is very sensitive to noise and outliers. The effectiveness of the proposed approach has been validated using synthetic examples and a benchmark structure.

• 천문학회보
• /
• 제37권2호
• /
• pp.108.2-108.2
• /
• 2012

Recent results from microlensing surveys show that a free-floating planet or a wide-separation planet is more numerous than a main-sequence star in the Galaxy. Moreover, the detection efficiency of the planets will be improved in next-generation experiments with a high survey monitoring frequency. However, microlensing events produced by both planets appear similar light curves with a short duration timescale, thus it is difficult to distinguish them. In this paper, we investigated the detectable separation range of a wide-separation planet as the planet bound to its host star. We construct the fractional deviation maps using the magnifications of the planetary lensing and the single-lensing by planet itself for various parameters such as a mass ratio, separation, and source radius. As a result, we found that the pattern of the fractional deviation is related to the ratio of source radius to caustic size, and the ratio satisfying the detection criterion (i.e., ${\geq}5%$ in the fractional deviation) varies with a separation. Hence, we derived a fitting formula as the function of a mass ratio and a source radius to reflect the variation in the calculations of the detectable separation range of a wide-separation planet as the planet bound to its host star. In addition, we estimated the condition that a wide-separation planet can be detected as a single-lensing event under the finite source effect. We found that such a case is possible provided that the source radius is smaller than ~2.5 times of Einstein ring radius of a planet, regardless of a separation or a mass ratio.

• The Journal of the Acoustical Society of Korea
• /
• 제21권4E호
• /
• pp.146-155
• /
• 2002

We present a new technique for achieving source separation when given only a single charmel recording. The main idea is based on exploiting the inherent time structure of sound sources by learning a priori sets of time-domain basis functions that encode the sources in a statistically efficient manner. We derive a learning algorithm using a maximum likelihood approach given the observed single charmel data and sets of basis functions. For each time point we infer the source parameters and their contribution factors. This inference is possible due to the prior knowledge of the basis functions and the associated coefficient densities. A flexible model for density estimation allows accurate modeling of the observation, and our experimental results exhibit a high level of separation performance for simulated mixtures as well as real environment recordings employing mixtures of two different sources. We show separation results of two music signals as well as the separation of two voice signals.

• 한국음향학회지
• /
• 제21권4호
• /
• pp.146-146
• /
• 2002

We present a new technique for achieving source separation when given only a single channel recording. The main idea is based on exploiting the inherent time structure of sound sources by learning a priori sets of time-domain basis functions that encode the sources in a statistically efficient manner. We derive a learning algorithm using a maximum likelihood approach given the observed single channel data and sets of basis functions. For each time point we infer the source parameters and their contribution factors. This inference is possible due to the prior knowledge of the basis functions and the associated coefficient densities. A flexible model for density estimation allows accurate modeling of the observation, and our experimental results exhibit a high level of separation performance for simulated mixtures as well as real environment recordings employing mixtures of two different sources. We show separation results of two music signals as well as the separation of two voice signals.

• 융합신호처리학회논문지
• /
• 제15권3호
• /
• pp.91-96
• /
• 2014

두 개의 입력소자에 의한 음원방향 및 분리방법으로서는 연산량이 적고, 음원분리 성능이 높은 주파수 양귀 모델(Frequency Domain Binaural Model : FDBM)이 있다. FDBM은 주파수 영역에서 양귀간 위상차(Interaural Phase Difference : IPD) 및 양귀간 레벨차(Interaural Level Difference : ILD)를 구하여 음향신호가 오는 방향과 음원의 분리처리를 한다. 그러나 실제 환경에서는 반사음의 문제가 되고 있다. 이러한 반사음에 의한 영향을 줄이기 위하여 선행음 효과에 의한 직접음의 음상정위를 모의하여 초기 도착음을 검출하고 직접음이 오는 방향과 음원분리 방법을 제시하였다. 제시한 방법을 이용하여 음원방향 추정 및 분리에 대한 성능을 시뮬레이션으로 검토하였다. 그 결과, 방향추정은 음원이 오는 방향에서 ${\pm}10%$의 범위로 집중되어 음원의 방향과 가까운 값으로 추정되었다, 반사음이 존재하는 경우의 음원분리는 기존의 FDBM에 비하여 코히런스(Coherence), 음성품질 지각평가 PESQ(Perceptual Evaluation of Speech Quality : PESQ)가 높고, 정면에서의 지향특성 감쇠량이 작아 분리의 정도가 개선됨을 나타내었다. 그러나 반사음이 존재하지 않는 경우는 분리 정도가 낮았다.

• 한국음향학회지
• /
• 제31권5호
• /
• pp.332-339
• /
• 2012

빔공간 변환(beamspace transform) 기법은 공간 영역의 신호를 입사각 혹은 그 사인함수의 영역으로 변환하는 기법으로, MUSIC과 같은 음원 정위 및 추적(source localization and tracking) 문제나 적응 빔형성(adaptive beamforming)과 같은 문제에서 많이 사용되는 기법이다. 다채널 음원 분리 기법에 사용될 때에는, 음원의 정보 뿐만아니라 해당 음원의 이미지(image)를 재구성하여야 하므로 역변환 기법 또한 중요하다. 본 논문에서는 멀티 채널 음원 분리 기법을 위한 빔공간 변환 기법과 그 역변환 기법에 대하여 고찰하였으며, 특히 빔공간-영역 다채널 비음수 행렬 분해 기법에 적용되었을 때 그 성능에 미치는 영향을 중점적으로 살펴보았다.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2002년도 봄 학술발표논문집 Vol.29 No.1 (B)
• /
• pp.289-291
• /
• 2002

Host of source separation methods focus on stationary sources so higher-order statistics is necessary In this paler we consider a problem of source separation when sources are second-order nonstationary stochastic processes . We employ the natural gradient method and develop learning algorithms for both 1inear feedback and feedforward neural networks. Thus our algorithms possess equivariant property Local stabi1iffy analysis shows that separating solutions are always locally stable stationary points of the proposed algorithms, regardless of probability distributions of