Low cost sparse subspace tracking algorithms

Lassami, Nacerredine; Bey, Abdeldjalil Aïssa El; Abed-Meraim, Karim

doi:10.1016/j.sigpro.2020.107522

Cited by 11 publications

(6 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In a more recent result [32], we have discussed the conditions to have a unique solution that corresponds to the sparse MIMO channel H. Without going deep into details, our results stated that if we have high level of sparsity of the MIMO channel H and a large number of receivers ((L + 1)N r > CN t log(N t )…”

Section: Initialization and Resolving Ambiguities In Subspace-based B...mentioning

confidence: 91%

“…with C is a positive constant related to the sparsity of the channel), then we can estimate the sparse channel H up to complex diagonal scaling matrix and permutation matrix. If these conditions are satisfied, we can use iteratively the second step of the algorithm SS-FAPI [32] (System matrix Sparsity based on Fast Approximated Power Iterations) in order to estimate the sparse channel matrix H from the subspace channel method estimation ĤSS . The SS-FAPI algorithm is an adaptive algorithm based on a two step approach, where the first one uses the FAPI algorithm [33] for the adaptive extraction of an orthonormal basis of the principal subspace equivalent to Ĥt SS (changes over the time t).…”

Section: Initialization and Resolving Ambiguities In Subspace-based B...mentioning

confidence: 99%

See 1 more Smart Citation

Blind joint MIMO channel and data estimation based on regularized ML

Lassami

Bey²,

Abed-Meraim³

2021

Digital Signal Processing

Self Cite

View full text Add to dashboard Cite

Section: Initialization and Resolving Ambiguities In Subspace-based B...mentioning

confidence: 91%

Section: Initialization and Resolving Ambiguities In Subspace-based B...mentioning

confidence: 99%

Blind joint MIMO channel and data estimation based on regularized ML

Lassami

Bey²,

Abed-Meraim³

2021

Digital Signal Processing

Self Cite

View full text Add to dashboard Cite

“…In this topic, we also proposed several two-stage approach based algorithms for sparse ST in [63,64,70]. The main steps of the two-stage approach is as follows.…”

Section: Sparse Subspace Trackingmentioning

confidence: 99%

“…We particularly exploited the natural gradient to find the sparsest matrix from the estimated orthonormal matrix by OPAST. In [64,70], we used FAPI in the first stage and then derived SS-FAPI, orthogonal SS-FAPI, and GSS-FAPI algorithms. Specifically, the sparsity criterion considered there is differentiable and smoother than the previous one in [63].…”

Section: Sparse Subspace Trackingmentioning

confidence: 99%

Robust Subspace Tracking Algorithms in Signal Processing: A Brief Survey

Thanh

Dũng

Trung

et al. 2021

REV J. Electron. Commun.

Self Cite

View full text Add to dashboard Cite

Principal component analysis (PCA) and subspace estimation (SE) are popular data analysis tools and used in a wide range of applications. The main interest in PCA/SE is for dimensionality reduction and low-rank approximation purposes. The emergence of big data streams have led to several essential issues for performing PCA/SE. Among them are (i) the size of such data streams increases over time, (ii) the underlying models may be time-dependent, and (iii) problem of dealing with the uncertainty and incompleteness in data. A robust variant of PCA/SE for such data streams, namely robust online PCA or robust subspace tracking (RST), has been introduced as a good alternative. The main goal of this paper is to provide a brief survey on recent RST algorithms in signal processing. Particularly, we begin this survey by introducing the basic ideas of the RST problem. Then, different aspects of RST are reviewed with respect to different kinds of non-Gaussian noises and sparse constraints. Our own contributions on this topic are also highlighted.

show abstract

“…[9]- [11]) have been recently presented. Here, data windows are not stored but processed as they are available, resulting in a complexity in the order of O(nk 2 ) or, at the price of some approximation, even O(nk).…”

Section: Introductionmentioning

confidence: 99%

Streaming Algorithms for Subspace Analysis: an Edge- and Hardware-oriented review

Marchioni¹,

Prono²,

Mangia³

et al. 2021

Preprint

View full text Add to dashboard Cite

Subspace analysis is a basic tool for coping with high-dimensional data and is becoming a fundamental step in early processing of many signals elaboration tasks. Nowadays trend of collecting huge quantities of usually very redundant data by means of decentralized systems suggests these techniques be deployed as close as possible to the data sources. Regrettably, despite its conceptual simplicity, subspace analysis is ultimately equivalent to eigenspace computation and brings along non-negligible computational and memory requirements. To make this fit into typical systems operating at the edge, specialized streaming algorithms have been recently devised that we here classify and review giving them a coherent description, highlighting features and analogies, and easing comparisons. Implementation of these methods is also tested on a common edge digital hardware platform to estimate not only abstract functional and complexity features, but also more practical running times and memory footprints on which compliance with real-world applications hinges.

show abstract

Low cost sparse subspace tracking algorithms

Cited by 11 publications

References 30 publications

Blind joint MIMO channel and data estimation based on regularized ML

Blind joint MIMO channel and data estimation based on regularized ML

Robust Subspace Tracking Algorithms in Signal Processing: A Brief Survey

Streaming Algorithms for Subspace Analysis: an Edge- and Hardware-oriented review

Contact Info

Product

Resources

About