Optimal design of three-dimensional filter for transmission channel represented by state-space model with uncertain parameters and orthogonal descriptor

• Published in: Multimedia tools and applications Vol. 83; no. 18; pp. 53975 - 53999
• Main Authors: Zouhri, Amal, Kririm, Said, El Mallahi, Mostafa, Hmamed, Abdelaziz
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.05.2024; Springer Nature B.V
• Subjects: Algorithms; Apexes; Computer Communication Networks; Computer Science; Data Structures and Information Theory; Deconvolution; Design optimization; Feature extraction; Filtration; H infinity; Linear matrix inequalities; Methods; Multimedia; Multimedia Information Systems; Neural networks; Noise; Object recognition; Parameter robustness; Parameter uncertainty; Performance evaluation; Performance indices; Robustness; Signal processing; Special Purpose and Application-Based Systems; State space models; Orthogonal descriptor; 3D Racah moment; PSNR; more; Roesser state-space model; Transmission channel; MSC code2; MSC code1; Features extraction; MSE; Polytopic uncertainties
• ISSN: 1573-7721; 1380-7501; 1573-7721
• DOI: 10.1007/s11042-023-16409-3

The proposed work focuses on a new set of robust H ∞ deconvolution filtering of 3-D objects using feature extraction from Orthogonal Descriptor such as Racah moment, and Roesser local state-space model with uncertain parameters. The time-invariant uncertain parameters are supposed to belong to a polytope with known vertices. The main idea is to design robust H ∞ deconvolution filter to reconstruct the noisy 3D object from the feature extractions of Racah moments. Furthermore, the filtering error system is asymptotically stable and satisfies the H ∞ performance index for all admissible uncertainties. The sufficient condition is given to ensure the H ∞ performance of the filtering error system through the parameter-dependent linear matrix inequalities (LMIs) constraints, and the Racah moment to give the feature extraction according to the order defined in advance instead of the global 3-D object. Moreover, the robust 3-D deconvolution filter is designed to achieve the H ∞ performance index which the robust filter parameters are determined with certain optimization resolution. Finally, simulation result is shown to demonstrate the usefulness of the suggested design approach.