Low-Complexity Feature Stochastic Gradient Algorithm for Block-Lowpass Systems

Yazdanpanah, Hamed; Diniz, Paulo S. R.; Lima, Markus V. S.

doi:10.1109/access.2019.2944187

Cited by 7 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are various candidates for

𝒫

to exploit revealed sparsity, such as the

{l}_0

‐norm, 13‐15 the

{l}_1

‐norm, 16‐19 the thresholding approaches, 20‐23 etc. In this work, we adopt the

{l}_1

‐norm due to its simplicity and computational efficiency.…”

Section: The If‐lms Algorithmmentioning

confidence: 99%

“…Once the hidden sparsity is revealed by F(k), the function  should exploit the exposed sparsity. There are various candidates for  to exploit revealed sparsity, such as the l 0 -norm, [13][14][15] the l 1 -norm, [16][17][18][19] the thresholding approaches, [20][21][22][23] etc. In this work, we adopt the l 1 -norm due to its simplicity and computational efficiency.…”

Section: The If-lms Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

Improved feature least mean square algorithm

Yazdanpanah

2022

Adaptive Control & Signal

Self Cite

View full text Add to dashboard Cite

In this paper, we propose the improved feature least-mean-square (IF-LMS) algorithm to exploit hidden sparsity in unknown systems. Recently, the feature least-mean-square (F-LMS) algorithm has been introduced, but its application is limited to particular systems since it uses predetermined feature matrices. However, the proposed IF-LMS algorithm utilizes the stochastic gradient descent (SGD) method to learn feature matrices; thus, it can be used in any system that the classical LMS algorithm is applicable. Hence, by employing a learnable feature matrix, the IF-LMS algorithm has a vast application area as compared to the F-LMS algorithm. Moreover, mathematically, we discuss some parameters of the IF-LMS algorithm. Simulation results, in synthetic and real-life scenarios, demonstrate that the IF-LMS algorithm has superior filtering accuracy to the well-known LMS algorithm.

show abstract

“…There are various candidates for

𝒫

to exploit revealed sparsity, such as the

{l}_0

‐norm, 13‐15 the

{l}_1

‐norm, 16‐19 the thresholding approaches, 20‐23 etc. In this work, we adopt the

{l}_1

‐norm due to its simplicity and computational efficiency.…”

Section: The If‐lms Algorithmmentioning

confidence: 99%

Section: The If-lms Algorithmmentioning

confidence: 99%

Improved feature least mean square algorithm

Yazdanpanah

2022

Adaptive Control & Signal

Self Cite

View full text Add to dashboard Cite

show abstract

“…, L. In other words, S k is a set of indices i that corresponds to g(i) = k in actual. Then a new (L + 1)-point sequence { a k (n) } is defined by subsets S k [28], which is…”

Section: Cross-correlationmentioning

confidence: 99%

The Algorithm and Structure for Digital Normalized Cross-Correlation by Using First-Order Moment

Pan

Xia

et al. 2020

Sensors

View full text Add to dashboard Cite

Normalized cross-correlation is an important mathematical tool in digital signal processing. This paper presents a new algorithm and its systolic structure for digital normalized cross-correlation, based on the statistical characteristic of inner-product. We first introduce a relationship between the inner-product in cross-correlation and a first-order moment. Then digital normalized cross-correlation is transformed into a new calculation formula that mainly includes a first-order moment. Finally, by using a fast algorithm for first-order moment, we can compute the first-order moment in this new formula rapidly, and thus develop a fast algorithm for normalized cross-correlation, which contributes to that arbitrary-length digital normalized cross-correlation being performed by a simple procedure and less multiplications. Furthermore, as the algorithm for the first-order moment can be implemented by systolic structure, we design a systolic array for normalized cross-correlation with a seldom multiplier, in order for its fast hardware implementation. The proposed algorithm and systolic array are also improved for reducing their addition complexity. The comparisons with some algorithms and structures have shown the performance of the proposed method.

show abstract

“…In this subsection, we utilize the AP, ST-SM-AP, and DT-SM-AP algorithms to identify a measured unknown system corresponding to the room impulse response (RIR) tested in [16,21]. For the AP algorithm, the step-size is chosen as µ = 0.9.…”

Section: Real-life Examplementioning

confidence: 99%

Data Censoring with Set-Membership Affine Projection Algorithm

Karamali¹,

Zardadi²,

Moradi³

2020

csci

View full text Add to dashboard Cite

In this paper, the set-membership affine projection (SM-AP) algorithm is utilized to censor non-informative data in big data applications. To this end, the probability distribution of the additive noise signal and the excess of the mean-squared error (EMSE) in steady-state are employed in order to estimate the threshold parameter of the single threshold SM-AP (ST-SM-AP) algorithm aiming at attaining the desired update rate. Furthermore, by defining an acceptable range for the error signal, the double threshold SM-AP (DT-SM-AP) algorithm is proposed to detect very large errors due to the irrelevant data such as outliers. The DT-SM-AP algorithm can censor non-informative and irrelevant data in big data applications, and it can improve the misalignment and convergence rate of the learning process with high computational efficiency. The simulation and numerical results corroborate the superiority of the proposed algorithms over traditional algorithms.

show abstract

Low-Complexity Feature Stochastic Gradient Algorithm for Block-Lowpass Systems

Cited by 7 publications

References 24 publications

Improved feature least mean square algorithm

Improved feature least mean square algorithm

The Algorithm and Structure for Digital Normalized Cross-Correlation by Using First-Order Moment

Data Censoring with Set-Membership Affine Projection Algorithm

Contact Info

Product

Resources

About