Sensor placement optimization of blind source separation in a wireless acoustic sensor network via hybrid descent methods

Wang, Qingzheng; Low, Siow Yong; Li, Zhibao; Yiu, Ka Fai Cedric

doi:10.1016/j.apacoust.2021.108509

Cited by 6 publications

(2 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ming et al [3] proposed an adaptive sparse denoising and periodicity weighted spectrum separation for obtaining features of composite defects in rolling bearings. When the source and signal channels are ambiguous, blind source separation (BSS) is a technique for separating the target signal of the observed signal [4]. For instance, Zhu et al [5] used empirical mode decomposition combined with principal component analysis (PCA) to conduct BSS for bearing composite faults.…”

Section: Introductionmentioning

confidence: 99%

Blind extraction of acoustic signals from rolling bearing composite faults based on ISCA-IMCKD

Jialu¹,

Zhou²,

Liu³

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

In the production practice, the signal gathered by sensor often includes strong ambient noise, and its composition is complex. Focusing on the trouble that traditional methods are difficult to separate and extract fault frequency from strong background noise, a novel compound fault blind extraction means based on improved sparse component analysis (ISCA) and improved maximum correlation kurtosis deconvolution (IMCKD)-named ISCA‐IMCKD- is suggested. Initially, the signal that the sensor has collected is shifted into time-frequency area signal by short-time Fourier transform (STFT). In addition, the single source domain characteristic data is screened by improved single source point detection (ISSPD) to decide the number of sources. Secondly, ISCA method is optimized by using cosine distance improved fuzzy C-means clustering, which is utilized to further process the characteristic data to calculate the mixing matrix. Moreover, estimated source signal is initially extracted according to the membership degree of clustering results. Finally, estimated source signal is shift into the time area by inverse STFT transform, and the IMCKD is employed to enhance the characteristics of the projected source signals. Meanwhile, initially estimated source signal is completely separated, and the defect frequency of the composite faults are finally extracted by envelope analysis. Simulation experiments and measured data are employed to certify the viability of the proposed means. The defect detection of rolling bearings is finished while the time cost is significantly saved.

show abstract

Section: Introductionmentioning

confidence: 99%

Blind extraction of acoustic signals from rolling bearing composite faults based on ISCA-IMCKD

Jialu¹,

Zhou²,

Liu³

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…The classic methods of beamformer design include the minimum variance distortionless response (MVDR) beamformer design, the linearly constrained minimum variance (LCMV) beamformer design [3,4], and the generalized sidelobe canceller (GSC) [5,6]. Generally, the performance of beamforming is greatly affected by both the filter coefficients and the array configuration [7,8,9,10]. Recently, the sparsity of array configuration attracts more and more attentions, and various strategies have been developed to devise sparse arrays for different tasks including direction finding, adaptive beamforming, and beampattern synthesis [11,12,13,14].…”

Section: Introductionmentioning

confidence: 99%

Sparse broadband beamformer design via proximal optimization Techniques

2023

J. Nonlinear Var. Anal.

View full text Add to dashboard Cite

Beamforming is one of the most important techniques to enhance the quality of signal in array sensor signal processing, and the performance of a beamformer is usually related to the design of array configuration and beamformer weight. Recently, it was realized that the sparsity of the filter coefficients can reduce the cost of signal acquisition and communication, and as a consequence, the sparse broadband beamformer design attracts more and more attentions. In this paper, we first propose a proximal sparse beamformer design model which obtains the sparse and robust filter coefficients through solving a composite optimization problem. The objective function of the model is the sum of a least squares term, a proximal term, and an 1 -regularization term. The least squares term reflects the data fidelity; the proximal term, whose center is predetermined via a simple least squares, enhances the robustness; while the 1 term ensures the sparsity of the solution. This model not only maintains the authenticity of the least squares solution, but also ensures the sparsity of the filter coefficients. A significant feature of the model is that we use 'partial' data to obtain the least squares solution and use another 'partial' data to construct the data fidelity term, which can evidently decrease the computational cost. For solving the composite optimization problem, we tailor several popular algorithms, such as the alternating direction method of multipliers, the forward-backward splitting method, and the Douglas-Rachford splitting method. Numerical results observably exhibit the improvements of the proposed approach over existing works in both effectiveness and efficiencies.

show abstract

Research on Acoustic Signal Denoising Algorithm Based on Blind Source Separation

Xue,

Xu,

2024

Procedia Computer Science

View full text Add to dashboard Cite

Sensor placement optimization of blind source separation in a wireless acoustic sensor network via hybrid descent methods

Cited by 6 publications

References 37 publications

Blind extraction of acoustic signals from rolling bearing composite faults based on ISCA-IMCKD

Blind extraction of acoustic signals from rolling bearing composite faults based on ISCA-IMCKD

Sparse broadband beamformer design via proximal optimization Techniques

Research on Acoustic Signal Denoising Algorithm Based on Blind Source Separation

Contact Info

Product

Resources

About