Recursive Least-Squares Algorithms for the Identification of Low-Rank Systems

Elisei-Iliescu, Camelia; Paleologu, Constantin; Benesty, Jacob; Stanciu, Cristian; Anghel, Cristian; Ciochină, Silviu

doi:10.1109/taslp.2019.2903276

Cited by 78 publications

(39 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In common acoustic applications, most of the linear systems are low-rank in nature due to the redundancies resulted from the room reflections and/or sparseness in the acoustic system. The impulse response of an acoustic channel can be approximated by a low-rank model that involves the nearest Kronecker product between a group of short impulse responses [10], [14]. As a consequence, we can also consider to decompose the adaptive filter using the Kronecker product as…”

Section: Signal Model and Optimization Criterionmentioning

confidence: 99%

“…In the RLS-NKP algorithm [10], the squared error is used to define the cost function, which yields an effective adaptive algorithm robust to Gaussian noise. The resulting algorithm, however, was found to have poor performance in terms of convergence in non-Gaussian noise.…”

Section: Signal Model and Optimization Criterionmentioning

confidence: 99%

“…Recently, a recursive least-squares algorithm based on the nearest Kronecker product (RLS-NKP) decomposition was proposed to identify low-rank systems [10]. In such a method, the impulse response of an acoustic system is decomposed based on the nearest Kronecker product and a low-rank approximation.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Robust Recursive Least M-Estimate Adaptive Filter for the Identification of Low-Rank Acoustic Systems

Chen

Benesty

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

To identify acoustic systems (which are low-rank in nature) in non-Gaussian and Gaussian noise, a robust recursive least M-estimate adaptive filtering algorithm is developed in this paper by applying the nearest Kronecker product to decompose the acoustic impulse response. Two M-estimators, i.e., the Cauchy and Welsch estimators, are employed to define the cost function of the adaptive filter, leading to a class of numerically stable adaptive filtering algorithms, which are robust to non-Gaussian noise. The effectiveness of the developed algorithm is validated in acoustic environments with both Gaussian and non-Gaussian noise.

show abstract

Section: Signal Model and Optimization Criterionmentioning

confidence: 99%

Section: Signal Model and Optimization Criterionmentioning

confidence: 99%

See 1 more Smart Citation

Robust Recursive Least M-Estimate Adaptive Filter for the Identification of Low-Rank Acoustic Systems

Chen

Benesty

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…In both cases, the partial cost functions from Equations (30) and (33) can be minimized with respect to h 2 and h 3 , respectively.…”

Section: Trilinear Wiener Filtermentioning

confidence: 99%

“…In perspective, it would be useful to extend this approach to identify more general forms of impulse responses. Recently, we developed such ideas in References [29,30], based on the Wiener filter and the recursive least-squares (RLS) algorithm, by exploiting the nearest Kronecker product decomposition and the related low-rank approximation.…”

Section: Introductionmentioning

confidence: 99%

System Identification Based on Tensor Decompositions: A Trilinear Approach

et al. 2019

Self Cite

View full text Add to dashboard Cite

The theory of nonlinear systems can currently be encountered in many important fields, while the nonlinear behavior of electronic systems and devices has been studied for a long time. However, a global approach for dealing with nonlinear systems does not exist and the methods to address this problem differ depending on the application and on the types of nonlinearities. An interesting category of nonlinear systems is one that can be regarded as an ensemble of (approximately) linear systems. Some popular examples in this context are nonlinear electronic devices (such as acoustic echo cancellers, which are used in applications for two-party or multi-party voice communications, e.g., videoconferencing), which can be modeled as a cascade of linear and nonlinear systems, similar to the Hammerstein model. Multiple-input/single-output (MISO) systems can also be regarded as separable multilinear systems and be treated using the appropriate methods. The high dimension of the parameter space in such problems can be addressed with methods based on tensor decompositions and modelling. In recent work, we focused on a particular type of multilinear structure—namely the bilinear form (i.e., two-dimensional decompositions)—in the framework of identifying spatiotemporal models. In this paper, we extend the work to the decomposition of more complex systems and we propose an iterative Wiener filter tailored for the identification of trilinear forms (where third-order tensors are involved), which can then be further extended to higher order multilinear structures. In addition, we derive the least-mean-square (LMS) and normalized LMS (NLMS) algorithms tailored for such trilinear forms. Simulations performed in the context of system identification (based on the MISO system approach) indicate the good performance of the proposed solution, as compared to conventional approaches.

show abstract

Real‐Time Update Algorithm of Optical Fiber Distribution Network Equipment Based on Optical Imaging Technology

Shen,

Tong,

Yang

et al. 2024

Micro & Optical Tech Letters

View full text Add to dashboard Cite

With the increasing digitalization of distribution network equipment (DNE), real‐time update algorithms for digital twin (DT) models have become a research focus on the digitalization of DNE. However, traditional real‐time update algorithms for DT models still have problems such as poor real‐time, accuracy, robustness, and scalability. To better promote the development of digitalization of DNE, this article aimed to study the real‐time update algorithm of DT models using the Internet of Things (IoT) and optical imaging technology, to achieve real‐time updates of DT models of DNE. The article first described the problems existing in the traditional DT model of DNE. Then, IoT sensors and optical devices were used to collect data related to DNE; the Savitzky–Golay filtering algorithm was used to denoise the data. This article combined the IoT and optical imaging technology to construct a DT model; using the recursive least squares method again, key parameters and state parameters were extracted from the constructed DT mechanism model, achieving real‐time updates of the DNE DT model. Finally, to verify the application effect of the IoT and optical imaging technology in real‐time update algorithms for DT models of DNE, this article compared them with traditional parameter sensitivity analysis and state estimation. The research results showed that in the real‐time and accuracy testing of test case 13, the algorithm used in this article had a time of 0.014 s and an accuracy of 93.2%. The parameter sensitivity analysis method had a time of 0.045 s and an accuracy of 80.4%. The state estimation method took 0.056 s and had an accuracy of 82.7%. In addition, the robustness and scalability of the real‐time update algorithm for the DNE DT model using the method proposed in this article were significantly better than the other two traditional methods. The results showed that the real‐time update algorithm of the DT model of DNE based on the IoT and optical imaging technology had better real‐time performance, higher accuracy, and better robustness and scalability. This study highlights the significant impact of the IoT and optical imaging technology on the accuracy, robustness, and real‐time performance of real‐time update algorithms for DT models. This provides more solutions for real‐time monitoring, prediction, and control of DNE.

show abstract

Recursive Least-Squares Algorithms for the Identification of Low-Rank Systems

Cited by 78 publications

References 54 publications

Robust Recursive Least M-Estimate Adaptive Filter for the Identification of Low-Rank Acoustic Systems

Robust Recursive Least M-Estimate Adaptive Filter for the Identification of Low-Rank Acoustic Systems

System Identification Based on Tensor Decompositions: A Trilinear Approach

Real‐Time Update Algorithm of Optical Fiber Distribution Network Equipment Based on Optical Imaging Technology

Contact Info

Product

Resources

About