Block-Sparsity Log-sum-Induced Adaptive Filter for Cluster Sparse System Identiﬁcation

Aihua Zhang; Pengcheng Liu; Jun Sun; Bing Ning

Block-Sparsity Log-sum-Induced Adaptive Filter for Cluster Sparse System Identiﬁcation

Aihua Zhang, Pengcheng Liu, Jun Sun, Bing Ning

Computer Science

Research output: Contribution to journal › Article › peer-review

Abstract

In this work, an effective adaptive block sparsity log-sum least mean square (BSLS-LMS) algorithm is proposed to improve the convergence performance of cluster sparse system identification. The main idea of the proposed scheme is to add a new block-sparsity induced term into the cost function of LMS algorithm. We utilize l_1 norm of adaptive tap weights and log-sum as a mixed constraint, via optimizing the cost function through the gradient descent method, the proposed adaptive filtering method can iteratively move the identified signals towards the optimal solutions, and finally identify the unknown system accurately. The cluster-sparse system response, with block length and arbitrary average sparsity, is generated by a Markov-Gaussian (M-G) model. For the white Gaussian input data, the theoretical formulas of steady-state mis-adjustment and convergence behaviors of BSLS-LMS are derived in general sparse system and block sparse system, respectively. Numerical experiments demonstrate that the proposed adaptive BSLS-LMS algorithm performs much better convergence behavior than conventional sparse signal recovery solutions. The experimental study also verifies the consistency between the simulations results and the theoretical analysis.

Original language	English
Journal	IEEE Access
Publication status	Accepted/In press - 17 Sept 2020

Bibliographical note

This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy. Further copying may not be permitted; contact the publisher for details.

Access to Document

Z251DYKQ_BSLS_LMS____final
This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy. Further copying may not be permitted; contact the publisher for details.
Accepted author manuscript, 174 KBLicence: Unspecified

Cite this

@article{a9a4169a417940d6903211fbe461a6c5,

title = "Block-Sparsity Log-sum-Induced Adaptive Filter for Cluster Sparse System Identiﬁcation",

abstract = "In this work, an effective adaptive block sparsity log-sum least mean square (BSLS-LMS) algorithm is proposed to improve the convergence performance of cluster sparse system identification. The main idea of the proposed scheme is to add a new block-sparsity induced term into the cost function of LMS algorithm. We utilize l_1 norm of adaptive tap weights and log-sum as a mixed constraint, via optimizing the cost function through the gradient descent method, the proposed adaptive filtering method can iteratively move the identified signals towards the optimal solutions, and finally identify the unknown system accurately. The cluster-sparse system response, with block length and arbitrary average sparsity, is generated by a Markov-Gaussian (M-G) model. For the white Gaussian input data, the theoretical formulas of steady-state mis-adjustment and convergence behaviors of BSLS-LMS are derived in general sparse system and block sparse system, respectively. Numerical experiments demonstrate that the proposed adaptive BSLS-LMS algorithm performs much better convergence behavior than conventional sparse signal recovery solutions. The experimental study also verifies the consistency between the simulations results and the theoretical analysis.",

author = "Aihua Zhang and Pengcheng Liu and Jun Sun and Bing Ning",

note = "This is an author-produced version of the published paper. Uploaded in accordance with the publisher{\textquoteright}s self-archiving policy. Further copying may not be permitted; contact the publisher for details.",

year = "2020",

month = sep,

day = "17",

language = "English",

journal = "IEEE Access",

issn = "2169-3536",

publisher = "IEEE",

}

TY - JOUR

T1 - Block-Sparsity Log-sum-Induced Adaptive Filter for Cluster Sparse System Identiﬁcation

AU - Zhang, Aihua

AU - Liu, Pengcheng

AU - Sun, Jun

AU - Ning, Bing

N1 - This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy. Further copying may not be permitted; contact the publisher for details.

PY - 2020/9/17

Y1 - 2020/9/17

N2 - In this work, an effective adaptive block sparsity log-sum least mean square (BSLS-LMS) algorithm is proposed to improve the convergence performance of cluster sparse system identification. The main idea of the proposed scheme is to add a new block-sparsity induced term into the cost function of LMS algorithm. We utilize l_1 norm of adaptive tap weights and log-sum as a mixed constraint, via optimizing the cost function through the gradient descent method, the proposed adaptive filtering method can iteratively move the identified signals towards the optimal solutions, and finally identify the unknown system accurately. The cluster-sparse system response, with block length and arbitrary average sparsity, is generated by a Markov-Gaussian (M-G) model. For the white Gaussian input data, the theoretical formulas of steady-state mis-adjustment and convergence behaviors of BSLS-LMS are derived in general sparse system and block sparse system, respectively. Numerical experiments demonstrate that the proposed adaptive BSLS-LMS algorithm performs much better convergence behavior than conventional sparse signal recovery solutions. The experimental study also verifies the consistency between the simulations results and the theoretical analysis.

AB - In this work, an effective adaptive block sparsity log-sum least mean square (BSLS-LMS) algorithm is proposed to improve the convergence performance of cluster sparse system identification. The main idea of the proposed scheme is to add a new block-sparsity induced term into the cost function of LMS algorithm. We utilize l_1 norm of adaptive tap weights and log-sum as a mixed constraint, via optimizing the cost function through the gradient descent method, the proposed adaptive filtering method can iteratively move the identified signals towards the optimal solutions, and finally identify the unknown system accurately. The cluster-sparse system response, with block length and arbitrary average sparsity, is generated by a Markov-Gaussian (M-G) model. For the white Gaussian input data, the theoretical formulas of steady-state mis-adjustment and convergence behaviors of BSLS-LMS are derived in general sparse system and block sparse system, respectively. Numerical experiments demonstrate that the proposed adaptive BSLS-LMS algorithm performs much better convergence behavior than conventional sparse signal recovery solutions. The experimental study also verifies the consistency between the simulations results and the theoretical analysis.

M3 - Article

SN - 2169-3536

JO - IEEE Access

JF - IEEE Access

ER -