Sparse Representation for Brain Signal Processing: A tutorial on methods and applications

Li, Yuanqing; Yu, Zhu Liang; Bi, Ning; Xu, Yong; Gu, Zhenghui; Амари, Шун-ичи

doi:10.1109/msp.2013.2296790

Cited by 74 publications

(33 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The objective of the sparse representation is to estimate the scalar coefficients, so that we can sparsely represent the test signal as a linear combination of few atoms of dictionary D [14]. The sparse representation of an input signal y can be obtained by performing l 0 norm minimization as follows: min s s 0 subject to y = Ds (8) l 0 norm optimization gives us the sparse representation but it is an NP-hard problem [2].…”

Section: Designing a Dictionary Matrixmentioning

confidence: 99%

“…In compressive sensing (CS), this sparsity idea was used and according to CS theory, any natural signal can be epitomized sparsely on definite constraints [5,8]. If the signal and an over-complete dictionary matrix is given, then the objective of the sparse representation is to compute the sparse coefficients, so that the signal can be represented as a sparse linear combination of atoms (columns) in dictionary [14]. If the dictionary matrix is designed from the best extracted feature of MI signal, it helps to overcome the issue of inter-personal and intra-personal variability, also enhances the processing speed and accuracy of the classifier.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Classification of Motor Imagery Based EEG Signals Using Sparsity Approach

Sreeja

Rabha

Samanta

et al. 2017

Intelligent Human Computer Interaction

View full text Add to dashboard Cite

Abstract. The advancement in brain-computer interface systems (BCIs) gives a new hope to people with special needs in restoring their independence. Since, BCIs using motor imagery (MI) rhythms provides high degree of freedom, it is been used for many real-time applications, especially for locked-in people. The available BCIs using MI-based EEG signals usually makes use of spatial filtering and powerful classification methods to attain better accuracy and performance. Inter-subject variability and speed of the classifier is still a issue in MI-based BCIs. To address the aforementioned issues, in this work, we propose a new classification method, spatial filtering based sparsity (SFS) approach for MIbased BCIs. The proposed method makes use of common spatial pattern (CSP) to spatially filter the MI signals. Then frequency bandpower and wavelet features from the spatially filtered signals are used to bulid two different over-complete dictionary matrix. This dictionary matrix helps to overcome the issue of inter-subject variability. Later, sparse representation based classification is carried out to classify the two-class MI signals. We analysed the performance of the proposed approach using publicly available MI dataset IVa from BCI competition III. The proposed SFS method provides better classification accuracy and runtime than the well-known support vector machine (SVM) and logistic regression (LR) classification methods. This SFS method can be further used to develop a real-time application for people with special needs.

show abstract

Section: Designing a Dictionary Matrixmentioning

confidence: 99%

mentioning

confidence: 99%

Classification of Motor Imagery Based EEG Signals Using Sparsity Approach

Sreeja

Rabha

Samanta

et al. 2017

Intelligent Human Computer Interaction

View full text Add to dashboard Cite

show abstract

“…Blind signal separation(BSS) is to separate or extract the source signal from hybrid signal received by the sensor array in case the transmission channel characteristic and real source signal are unknown [1] [2]. It is one of the hottest new disciplines of signal processing and is widely applied to fields like communications, remote sensing, radar, sonar, and noise control.…”

Section: Introductionmentioning

confidence: 99%

Blind Signal Separation of Unknown Source Number under the Constraint Based on Uniform Linear Array

Huang¹,

Liao²,

Chen³

2017

Proceedings of the 2017 International Conference on Computational Science and Engineering (ICCSE 2017)

View full text Add to dashboard Cite

Abstract. Regarding blind signal separation based on eigenvalue decomposition, when the spatial dimension of signal is wrong in case of unknown source quantity, it will result in significant separation errors. By making use of kurtosis as the cost function, this paper attempts to construct a blind signal separation algorithm taking "space-kurtosis" spectrum as the basis, thus avoiding eigenvalue decomposition. Meanwhile, in the operation process, the source signal statistic independence and spatial distribution independence are fully utilized. It is proved by simulation experiments that the algorithm is with characteristics of high accuracy, fast operation and strong robustness.

show abstract

“…It has successfully extended from theoretical research to a variety of applications, such as signal classification [1], image and signal denoising [2,[31][32][33][34], blind sources separation [3] and so on. So far, most denosing literatures about sparse representation are image denoising [4]- [6], and signal denoising is rarely studied.…”

Section: Introductionmentioning

confidence: 99%

An Denoising Method Based on Analysis K-SVD and Disagreement Segment and Its Application on EMI Signal

Gong

Zhao

2017

Proceedings of the 3rd International Conference on Wireless Communication and Sensor Networks (WCSN 2016)

View full text Add to dashboard Cite

Abstract-Sparse representation plays an important role in signal processing. Recently, the analysis sparse representation has been attracting more and more attention. In this paper, an improved analysis K-SVD denoising algorithm based on disagreement-segment is proposed. A signal is first divided into small redundant segments, and then the signal can be denoised by the analysis K-SVD algorithm. Considering the gap between the local processing and the global signal recovery, we define a disagreement-segment as the difference between the intermediate locally denoised segment and its corresponding part in the final denoised signal. By adding the disagreementsegment to the analysis K-SVD algorithm, the denoising effect of the analysis K-SVD algorithm has a significant improvement. In addition, the experimental results show that the proposed method outperforms the analysis K-SVD algorithm and other advanced methods.

show abstract

Sparse Representation for Brain Signal Processing: A tutorial on methods and applications

Cited by 74 publications

References 30 publications

Classification of Motor Imagery Based EEG Signals Using Sparsity Approach

Classification of Motor Imagery Based EEG Signals Using Sparsity Approach

Blind Signal Separation of Unknown Source Number under the Constraint Based on Uniform Linear Array

An Denoising Method Based on Analysis K-SVD and Disagreement Segment and Its Application on EMI Signal

Contact Info

Product

Resources

About