Youcef Amine Medjaouri
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Youcef Amine Medjaouri has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
Achour Ales
2 shared publications- Dries Peumans
Department of Fundamental Electricity and Instrumentation, Vrije Universiteit Brussel, Brussels, Belgium
2 shared publications Badreddine Rekioua
Department of Electrical Engineering, Ecole Militaire Polytechnique, Algiers, Algeria
1 shared publicationHoussem Habbouche
1 shared publication- Johan Gyselinck
BEAMS, Universit libre de Bruxelles, Brussels, Belgium
1 shared publication Karim Benzaoui
1 shared publication
Latest Publications
- Open Access
- 9 October 2026
- Published by Elsevier in SSRN Electronic Journal
An Adaptive Neuro-Fuzzy Inference System for Retrieving the Complex Permittivity and Permeability of Composite Materials
Electromagnetic (EM) characterization of materials consists of retrieving their complex permittivity ε and permeability µ from measured scattering parameters (S-parameters), which describe how a material responds to EM waves. Conventional extraction methods, such as the Nicolson–Ross–Weir (NRW) algorithm, rely on analytical inversion and require strict experimental conditions, precise calibration, and careful reference-plane adjustment. Although widely adopted, these approaches remain sensitive to measurement uncertainties and can be computationally demanding. From an engineering standpoint, this paper proposes a noniterative alternative to NRW that avoids its calibration constraints and iterative processing overhead. From an artificial intelligence (AI) approach, an Adaptive Neuro-Fuzzy Inference System (ANFIS) is introduced to learn the intrinsic nonlinear mapping between measured Sparameters and the corresponding complex EM parameters. A large and diverse dataset is constructed by combining experimental measurements with full-wave numerical simulations, ensuring broad coverage of material behaviors. The ANFIS model is trained on this dataset and its accuracy is assessed through a systematic error analysis comparing predicted and reference values. Its performance is compared against three established machine-learning models, Artificial Neural Networks (ANN), Support Vector Regression (SVR), and eXtreme Gradient Boosting (XGBoost), demonstrating superior precision and stability. To validate the approach experimentally, S-parameters of a Teflon sample measured in the X-band are fed into the trained models, and the retrieved ε and µ are compared with reference values from the literature, confirming the reliability of the proposed method.
- 9 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessAn Adaptive Neuro-Fuzzy Inference System for Retrieving the Complex Permittivity and Permeability of Composite Materials
Electromagnetic (EM) characterization of materials consists of retrieving their complex permittivity ε and permeability µ from measured scattering parameters (S-parameters), which describe how a material responds to EM waves. Conventional extraction methods, such as the Nicolson–Ross–Weir (NRW) algorithm, rely on analytical inversion and require strict experimental conditions, precise calibration, and careful reference-plane adjustment. Although widely adopted, these approaches remain sensitive to measurement uncertainties and can be computationally demanding. From an engineering standpoint, this paper proposes a noniterative alternative to NRW that avoids its calibration constraints and iterative processing overhead. From an artificial intelligence (AI) approach, an Adaptive Neuro-Fuzzy Inference System (ANFIS) is introduced to learn the intrinsic nonlinear mapping between measured Sparameters and the corresponding complex EM parameters. A large and diverse dataset is constructed by combining experimental measurements with full-wave numerical simulations, ensuring broad coverage of material behaviors. The ANFIS model is trained on this dataset and its accuracy is assessed through a systematic error analysis comparing predicted and reference values. Its performance is compared against three established machine-learning models, Artificial Neural Networks (ANN), Support Vector Regression (SVR), and eXtreme Gradient Boosting (XGBoost), demonstrating superior precision and stability. To validate the approach experimentally, S-parameters of a Teflon sample measured in the X-band are fed into the trained models, and the retrieved ε and µ are compared with reference values from the literature, confirming the reliability of the proposed method.
- 8 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dielectrics and Electrical Insulation
- Vol. 32 (4) , 1923-1929
Equivalent Modeling of Multilayered Conductive Composite Materials for EMI Shielding Applications
This paper develops a single-layer equivalent model for multilayered shielding materials using measurements of complex permittivity and permeability. Through rectangular waveguide Transmission/Reflection measurements, electromagnetic parameters and shielding effectiveness (SE) are evaluated. The aim is to establish an equivalent single-layer structure matching the SE of the original multilayered structure. Initial validation involves testing on simulated and experimental data of a reference single-layer material. The methodology is then applied to a novel three-layer PANI-Fe3O4/CNT/PANI-Fe3O4 composite material. Unlike conventional approaches requiring new multilayered material synthesis, this approach combines existing and widely available materials, simplifying characterization and reducing cost and complexity to achieve the desired SE. Comparative analysis with analytical SE values from Transmission Line Matrix (TLM) formulation is conducted using simulated data. The proposed approach is validated within the X-band frequency range using analytical TLM formulation, numerical simulation, and experimental data. The proposed approach exhibits good performance in reconstructing the equivalent model.
- 8 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dielectrics and Electrical Insulation
- Vol. 32 (4) , 1923-1929
Equivalent Modeling of Multilayered Conductive Composite Materials for EMI Shielding Applications
This paper develops a single-layer equivalent model for multilayered shielding materials using measurements of complex permittivity and permeability. Through rectangular waveguide Transmission/Reflection measurements, electromagnetic parameters and shielding effectiveness (SE) are evaluated. The aim is to establish an equivalent single-layer structure matching the SE of the original multilayered structure. Initial validation involves testing on simulated and experimental data of a reference single-layer material. The methodology is then applied to a novel three-layer PANI-Fe3O4/CNT/PANI-Fe3O4 composite material. Unlike conventional approaches requiring new multilayered material synthesis, this approach combines existing and widely available materials, simplifying characterization and reducing cost and complexity to achieve the desired SE. Comparative analysis with analytical SE values from Transmission Line Matrix (TLM) formulation is conducted using simulated data. The proposed approach is validated within the X-band frequency range using analytical TLM formulation, numerical simulation, and experimental data. The proposed approach exhibits good performance in reconstructing the equivalent model.