An Online Scheme for Delayed MISO System Identification
Open accessAbstract
The issue of parametric estimation in time delay models is the main topic of this research article. Multiple-Input Single-Output (MISO) Continuous-Time (CT) systems with numerous unknown time delays characterize these models. Two different recursive parametric estimate techniques are explored in this paper, the strategic application of Sequential Nonlinear Least Squares (SNLS) to attain global convergence and the Recursive Least Squares (RLS) technique in conjunction with the Gauss-Newton algorithm with the goal of achieving local optimization. Both approaches contribute to the comprehensive understanding of the parametric estimation landscape for time delay models. In a pivotal stride towards enhancing convergence, the research proposes a hybridization of the two methods. This synergistic approach is designed to leverage the strengths of both SNLS and the RLS-Gauss-Newton combination, fostering improved overall convergence properties. To substantiate the credibility and effectiveness of the proposed methodologies, the conducted research provides comprehensive simulation results. These simulations offer concrete examples of the efficacy and practicality of the suggested techniques in real-world situations, and they make significant contributions to the field of parametric estimation for time delay models.
AI snapshot
Written from the paper’s full text.
TL;DR
The paper studies online parametric estimation for Multiple-Input Single-Output (MISO) continuous-time systems with multiple unknown time delays, comparing Sequential Nonlinear Least Squares (SNLS), Recursive Least Squares with Gauss-Newton (GN-RLS), and a hybrid of the two. Simulation results indicate the hybrid method estimates parameters and delays more accurately than either individual method.
Key findings
- The proposed hybrid GN-RLS-SNLS method is reported to surpass both SNLS and GN-RLS in simulation, yielding “more satisfactory results” and better capturing the simulated system’s intricacies.
- Table I reports estimates close to true values: for example, \(f_1\) true = 3, GN-RLS = 3.0310±0.0310, hybrid = 3.0094±0.0094; \(d_1\) true = 8.84, GN-RLS = 8.8690±0.029, hybrid = 8.8455±0.0055; \(d_2\) true = 1.87, GN-RLS = 1.8763±0.0063, hybrid = 1.8760±0.006.
- The paper states that GN-RLS estimates closely align with target values, suggesting high accuracy and precision, while the hybrid method’s numerical simulation outcomes surpass both SNLS and GN-RLS.
Methods
- The MISO continuous-time model includes multiple unknown time delays, parameterized rational functions \(F(s)\) and \(G_j(s)\), a low-pass filter \(K(s)\), and a bilinear transformation to obtain a discrete-time estimation model.
- Three identification approaches are explored: SNLS for simultaneous estimation of parameters and time delays; RLS for linear parameters combined with Gauss-Newton for nonlinear time-delay estimation; and a hybrid method combining GN-RLS and SNLS.
- Simulation settings include forgetting factor \(\beta = 0.882\), \(p_0 = 10^6\), and SNR = 5 dB; Table I compares true values, GN-RLS estimates, and hybrid estimates.
Limitations
- The paper does not explicitly state limitations.
- Evident from the text: validation is simulation-only, with no experimental or real-world data reported.
- Evident from the text: performance claims rely on one numerical example and Table I, with no quantitative error metrics, convergence rates, or statistical comparisons beyond the reported estimates.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
- Findingown data
In the numerical simulation, the proposed hybrid method outperforms both the SNLS and the GN-RLS methods.
“The outcomes of the numerical simulation conducted using the hybrid method presented in this study surpass the performance of both the SNLS and the GN-RLS methods, suggesting that the hybrid approach yields more satisfactory results, indicating a superior capability in capturing the intricacies of the simulated system or phenomenon.”
Quote in the paperSubstantial claim
- May combine two claims. 3
- Classified as probably not a substantial claim. 1
May combine two claims.
Findingown dataThe parameter estimates obtained by the GN-RLS method closely align with the target parameter values, indicating high accuracy and precision (Table I).
“Table I provides an obvious indication that the estimated parameters acquired through the GN-RLS method closely align with the target values, suggesting a high degree of accuracy and precision in the parameter estimations and highlighting the efficacy of the GN-RLS method in capturing the underlying characteristics of the data.”
Quote in the paperSubstantial claimMay combine two claims.
Interpretationown dataHybridizing SNLS with the RLS-Gauss-Newton combination leverages the strengths of both and improves overall convergence properties.
“This synergistic approach is designed to leverage the strengths of both SNLS and the RLS-Gauss-Newton combination, fostering improved overall convergence properties.”
Quote in the paperSubstantial claimMay combine two claims.
Interpretationown dataThe proposed identification techniques are efficacious and practical in real-world situations.
“These simulations offer concrete examples of the efficacy and practicality of the suggested techniques in real-world situations, and they make significant contributions to the field of parametric estimation for time delay models.”
Quote in the paperSubstantial claimClassified as probably not a substantial claim.
Study design and conductnone givenIn the hybrid-method simulation, the noise magnitude was adjusted so that the signal-to-noise ratio matched that used in the GN-RLS and SNLS simulations.
“In this simulation case, the noise magnitude is adjusted to get the same SNR as in the GN-RLS and the SNLS methods.”
Quote in the paperSubstantial claim
The paper relies on this without support.
Problems still exist in the identification of Multiple-Input Single-Output time delay systems.
“Even if the aforementioned advancements take place, problems still exist, particularly with the Multiple-Input Single-Output (MISO) time delay system identification.”
The paper relies on this without support.
The interaction of numerous inputs in a MISO system poses a unique set of challenges that requires innovative approaches.
“The interaction of numerous inputs poses a unique set of challenges, necessitating the emergence of innovative approaches.”
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Table I assigns two different true values to f1
MinorTable I lists f1 twice, once with a true value of 3 and once with a true value of 2. One of the parameter labels appears to be incorrect.
“f1 3 0 3.0310±0.03103.0094±0.0094”
“f1 2 0 2.0112±0.01121.9819±0.0181”
Cites
Least squares based and gradient based iterative identification for Wiener nonlinear systems
Wang, D., & Ding, F. (2010) · Signal Processing · doi:10.1016/j.sigpro.2010.11.004 · Cited by 226 · cited 1× · see citing works
“The generalized observation vector is: where ( ) ˆ j e k d and ˆ ( , ) k are described in [13].”
Gradient based and least squares based iterative algorithms for matrix equations AXB+CXTD=F
Xie, L., Liu, Y., & Yang, H. (2010) · Applied Mathematics and Computation · doi:10.1016/j.amc.2010.07.019 · Cited by 185 · see citing works
Digital parameter-adaptive control of processes with unknown dead time
Kurz, H., & Goedecke, W. (1981) · Automatica · doi:10.1016/0005-1098(81)90099-6 · Cited by 166 · see citing works
VERRIEST, E. I., & PEPE, P. (2009) · Lecture notes in control and information sciences · doi:10.1007/978-3-642-02897-7 · Cited by 104 · see citing works
Identification of time delays using a polynomial identification method
Gawthrop, P. J., & Nihtilä, M. T. (1985) · Systems & Control Letters · doi:10.1016/0167-6911(85)90020-9 · Cited by 103 · see citing works
Time Delay Systems: Methods, Applications and New Trends
Avdonin, S., & Pandolfi, L. (2012) · Lecture notes in control and information sciences · doi:10.1007/978-3-642-25221-1 · Cited by 89 · cited 1× · see citing works
“Relevant studies have been conducted in the fields of biology, chemistry, economics, mechanics, viscoelasticity, physics, physiology, population dynamics, and engineering sciences [1-8].”
BackgroundAdaptive time delay neural network structures for nonlinear system identification
Yazdizadeh, A., & Khorasani, K. (2002) · Neurocomputing · doi:10.1016/s0925-2312(01)00589-6 · Cited by 80 · see citing works
An asymptotic modal observer for linear autonomous time lag systems
Leyva-Ramos, J., & Pearson, A. E. (1995) · IEEE Transactions on Automatic Control · doi:10.1109/9.400473 · Cited by 79 · see citing works
Identification and U‐control of a state‐space system with time‐delay
Gu, Y., Zhu, Q., & Nouri, H. (2021) · International Journal of Adaptive Control and Signal Processing · doi:10.1002/acs.3345 · Cited by 79 · cited 1× · see citing works
“In the expansive landscape of system identification involving time delays, many techniques have been proposed [9-18].”
BackgroundBao, B., Xu, Y., Sheng, J., et al. (2010) · Mathematical and Computer Modelling · doi:10.1016/j.mcm.2010.12.034 · Cited by 70 · see citing works
Adaptive identification of linear time‐delay systems
Orlov, Y. V., Belkoura, L., Richard, J., & Dambrine, M. (2003) · International Journal of Robust and Nonlinear Control · doi:10.1002/rnc.850 · Cited by 64 · cited 1× · see citing works
“Relevant studies have been conducted in the fields of biology, chemistry, economics, mechanics, viscoelasticity, physics, physiology, population dynamics, and engineering sciences [1-8].”
BackgroundSimultaneous online identification of rational dynamics and time delay: a correlation-based approach
Rad, A. B., Lo, W. L., & Tsang, K. M. (2003) · IEEE Transactions on Control Systems Technology · doi:10.1109/tcst.2003.819594 · Cited by 49 · cited 1× · see citing works
“In the expansive landscape of system identification involving time delays, many techniques have been proposed [9-18].”
BackgroundAn indirect approach for online identification of continuous time‐delay systems
Sharma, S., & Padhy, P. K. (2021) · International Journal of Numerical Modelling Electronic Networks Devices and Fields · doi:10.1002/jnm.2947 · Cited by 20 · see citing works
An adaptive generalized predictive control of time-varying delay system
Gao, W., Zhou, M., Li, Y., et al. (2005) · doi:10.1109/icmlc.2004.1382309 · Cited by 16 · see citing works
Identifiability and Identification of Linear Systems with Delays
Belkoura, L., Dambrine, M., Orlov, Y., et al. (2004) · Lecture notes in computational science and engineering · doi:10.1007/978-3-642-18482-6_9 · Cited by 12 · see citing works
Identification of delayed system using instrumental variable method
Kaur, D., & Dewan, L. (2012) · Journal of Control Theory and Applications · doi:10.1007/s11768-012-0289-2 · Cited by 8 · see citing works
Identification of time delays in linear stochastic systems
Chen, R., & Loparo, K. A. (1993) · International Journal of Control · doi:10.1080/00207179308934447 · Cited by 8 · see citing works
A novel identification method for time delay processes
Majhi, S., & Atherton, D. P. (1999) · European Control Conference (ECC) · doi:10.23919/ecc.1999.7099664 · Cited by 2 · see citing works
Flagged first, then by citations
- Authors
- Ghoul, Y., Zitouni, N., & Flah, A.
- Affiliations
- National Engineering School of Tunis (TN)
- Tunis El Manar University (TN)
- Applied Science Private University (JO)
- Corresponding author
- Flah, Aymen
- Venue
- Engineering, Technology & Applied Science Research
- Published
- 2024-08-01
- Volume
- 14
- Issue
- 3
- Pages
- 14022-14027
- ISSN
- 1792-8036 (ISSN-L)
- 2241-4487
- OpenAlex
- W4402402282
- Cited by
- 0
- References
- 18
- FWCI
- 0.00
- Type
- journal article
- Language
- en
- Open access
- gold
- dx.doi.org
- Locations
- Engineering Technology & Applied Science Research (publishedVersion, cc-by)
- Topics
- Physical Sciences › Engineering › Control and Systems Engineering › Control Systems and Identification
- Physical Sciences › Engineering › Computational Mechanics › Advanced Adaptive Filtering Techniques
- Physical Sciences › Mathematics › Applied Mathematics › Statistical and numerical algorithms
- Keywords
- time-delay systems, MISO systems, continuous-time systems, parameter estimation, recursive least squares, Gauss-Newton method, hybrid estimation algorithm, global convergence, local optimization, online system identification
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-17.
Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-17.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
