• The Journal of the Korea Contents Association
• /
• v.6 no.12
• /
• pp.89-95
• /
• 2006

The performance of blind source separation(BSS) using independent component analysis (ICA) declines significantly in a reverberant environment. A post-processing method proposed in this paper was designed to remove the residual component precisely. The proposed method used modified NLMS(normalized least mean square) filter in frequency domain, to estimate cross-talk path that causes residual cross-talk components. Residual cross-talk components in one channel is correspond to direct components in another channel. Therefore, we can estimate cross-talk path using another channel input signals from adaptive filter. Step size is normalized by input signal power in conventional NLMS filter, but it is normalized by sum of input signal power and error signal power in modified NLMS filter. By using this method, we can prevent misadjustment of filter weights. The estimated residual cross-talk components are subtracted by non-stationary spectral subtraction. The computer simulation results using speech signals show that the proposed method improves the noise reduction ratio(NRR) by approximately 3dB on conventional FDICA.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.471-471
• /
• 2000

This paper presents a new kurtosis-based algorithm for blind separation of convolutively mixed source signals. The algorithm whitens the signals not only spatially but also temporally beforehand. A separator is built for the whitened signals and it exists in the set of para-unitary matrices. Since the set forms a curved manifold, it is hard to treat its elements. In order to avoid the difficulty, this paper introduces the Cayley transformation for the para-unitary matrices. The transformed matrix is referred to as para-skew-Hermitian matrix and the set of such matrices forms a linear space. In the set of all para-skew-Hermitian matrices, the kurtosis-based algorithm obtains a desired separator. This paper also shows the algorithm's application to electrogastrogram datum which are observed by 4 electrodes on subjects' abdomen around their stomachs. An electrogastrogram contains signals from a stomach and other organs. This paper obtains independent components by the algorithm and then extracts the signal corresponding to the stomach from the data.

• The Journal of the Acoustical Society of Korea
• /
• v.31 no.5
• /
• pp.317-331
• /
• 2012

In this paper, we develop a multichannel blind source separation algorithm based on a beamspace transform and the multichannel non-negative matrix factorization (NMF) method. The NMF algorithm is a famous algorithm which is used to solve the source separation problems. In this paper, we consider a beamspace-time-frequency domain data model for multichannel NMF method, and enhance the conventional method using a beamspace transform. Our decomposition algorithm is applied to audio source separation, using a dataset from the international Signal Separation Evaluation Campaign 2010 (SiSEC 2010) for evaluation.

• Proceedings of the IEEK Conference
• /
• 2000.09a
• /
• pp.609-612
• /
• 2000

본 논문에서는 음원에 관한 정보가 미지의 상황에서 마이크로폰 어레이를 사용하여 두 음원신호를 분리하는 ,시스템을 제안한다 이 시스템은 두 단계로 구성되어 있으며, 첫 번째 단계에서는 파워가 큰 제 1음원의 DOA(Direction Of Arrival)를 추정하고, AMUSE(Algorithm for Multiple Unknown Signals Extraction)법을 사용한 Blind Deconvolution에 의해 음원신호의 분리를 행한다 두 번째 단계에서는 파워가 낮은 제 2음원의 강조신호를 사용하여 DSA(Delay and Sum Array)법에 의해 제 2음원의 DOA를 추정하고,AMUSE법의 출력신호와 두 음원의 DOA를 이용하여 ANF(Adaptive Notch Filter)를 구성하고, 두 음원신호의 재 분리를 행한다. 그리고, 시뮬레이션을 통해 제안한 방법의 유효성을 검토한 결과 두 음원 신호가 분리 가능한 것이 확인되었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.10
• /
• pp.4814-4834
• /
• 2018

This paper presents an architecture for wireless sensor networks (WSNs) with blind source separation (BSS) applied to retrieve the received mixing signals of the sink nodes first. The little-to-no need of prior knowledge about the source signals of the sink nodes in the BSS method is obviously advantageous for WSNs. The optimization problem of the BSS of multiple independent source signals with complex and noncircular distributions from observed sensor nodes is considered and addressed. This paper applies Castella's reference-based scheme to Novey's negentropy-based algorithms, and then proposes a novel fast fixed-point (FastICA) algorithm, defined as the reference-signal negentropy complex FastICA (RSNT-cFastICA) for complex-valued noncircular-distribution source signals. The proposed method for the sink nodes is substantially more efficient than Novey's quasi-Newton algorithm in terms of computational speed under large numbers of samples, can effectively improve the power consumption effeciency of the sink nodes, and is significantly beneficial for WSNs and wireless communication networks (WCNs). The effectiveness and performance of the proposed method are validated and compared with three related BSS algorithms through theoretical analysis and simulations.

• Journal of KIISE:Software and Applications
• /
• v.36 no.3
• /
• pp.179-186
• /
• 2009

This paper proposes a target speech extraction which restores speech signal of a target speaker form noisy convolutive mixture of speech and an interference source. We assume that the target speaker is known and his/her utterances are available in the training time. Incorporating the additional information extracted from the training utterances into the separation, we combine convolutive blind source separation(CBSS) and non-negative decomposition techniques, e.g., probabilistic latent variable model. The nonnegative decomposition is used to learn a set of bases from the spectrogram of the training utterances, where the bases represent the spectral information corresponding to the target speaker. Based on the learned spectral bases, our method provides two postprocessing steps for CBSS. Channel selection step finds a desirable output channel from CBSS, which dominantly contains the target speech. Reconstruct step recovers the original spectrogram of the target speech from the selected output channel so that the remained interference source and background noise are suppressed. Experimental results show that our method substantially improves the separation results of CBSS and, as a result, successfully recovers the target speech.

• Geosystem Engineering
• /
• v.21 no.5
• /
• pp.251-261
• /
• 2018

With the urbanization in recent years, the power line interference noise in electromagnetic signal is increasing day by day, and has gradually become an unavoidable component of noises in magnetotelluric signal detection. Therefore, a kind of power line interference noise elimination method based on independent component analysis in wavelet domain for magnetotelluric signal is put forward in this paper. The method first uses wavelet decomposition to change single-channel signal into multi-channel signal, and then takes advantage of blind source separation principle of independent component analysis to eliminate power line interference noise. There is no need to choose the layer number of wavelet decomposition and the wavelet base of wavelet decomposition according to the observed signal. On the treatment effect, it is better than the previous power line interference removal method based on independent component analysis. Through the de-noising processing to actual magnetotelluric measuring data, it is shown that this method makes both the apparent resistivity curve near 50 Hz and the phase curve near 50 Hz become smoother and steadier than before processing, i.e., it effectively eliminates the power line interference noise.

• Phonetics and Speech Sciences
• /
• v.4 no.4
• /
• pp.87-92
• /
• 2012

Speech enhancement aims to improve speech quality by removing background noise from noisy speech. Independent vector analysis is a type of frequency-domain independent component analysis method that is known to be free from the frequency bin permutation problem in the process of blind source separation from multi-channel inputs. This paper proposed a new method of microphone array based speech enhancement that combines independent vector analysis and beamforming techniques. Independent vector analysis is used to separate speech and noise components from multi-channel noisy speech, and delay-sum beamforming is used to determine the enhanced speech among the separated signals. To verify the effectiveness of the proposed method, experiments for computer simulated multi-channel noisy speech with various signal-to-noise ratios were carried out, and both PESQ and output signal-to-noise ratio were obtained as objective speech quality measures. Experimental results have shown that the proposed method is superior to the conventional microphone array based noise removal approach like GSC beamforming in the speech enhancement.

• ETRI Journal
• /
• v.41 no.5
• /
• pp.658-669
• /
• 2019

For dual-channel time-frequency (TF) overlapped signals with low sparsity in underdetermined blind source separation (UBSS), this paper proposes an effective method based on interval probability to estimate and expand the types of mixing matrices. First, the detection of TF single-source points (TF-SSP) is used to improve the TF sparsity of each source. For more distinguishability, as the ratios of the coefficients from different columns of the mixing matrix are close, a local peak-detection mechanism based on interval probability (LPIP) is proposed. LPIP utilizes uniform subintervals to optimize and classify the TF coefficient ratios of the detected TF-SSP effectively in the case of a high level of TF overlap among sources and reduces the TF interference points and redundant signal features greatly to enhance the estimation accuracy. The simulation results show that under both noiseless and noisy cases, the proposed method performs better than the selected mainstream traditional methods, has good robustness, and has low algorithm complexity.

• Journal of Communications and Networks
• /
• v.14 no.5
• /
• pp.481-486
• /
• 2012

Accurate and efficient estimation of the number of sources is critical for providing the parameter of targets in problems of array signal processing and blind source separation among other such problems. When conventional estimators work in unfavorable scenarios, e.g., at low signal-to-noise ratio (SNR), with a small number of snapshots, or for sources with a different strength, it is challenging to maintain good performance. In this paper, the detection limit of the minimum description length (MDL) estimator and the signal strength required for reliable detection are first discussed. Though a comparison, we analyze the reason that performances of classical estimators deteriorate completely in unfavorable scenarios. After discussing the limiting distribution of eigenvalues of the sample covariance matrix, we propose a new approach for estimating the number of sources which is based on a sequential hypothesis test. The new estimator performs better in unfavorable scenarios and is consistent in the traditional asymptotic sense. Finally, numerical evaluations indicate that the proposed estimator performs well when compared with other traditional estimators at low SNR and in the finite sample size case, especially when weak signals are superimposed on the strong signals.