Durrmeyer sampling-type operators: approximation in variation

From MaRDI portal





The paper under review investigates the approximation properties of Durrmeyer sampling-type operators within the framework of functions of bounded variation. These operators combine classical continuous approximation methods, specifically the Durrmeyer modification of Bernstein polynomials, with discrete generalized sampling processes.\N\NThe authors focus on analytical properties in the space of functions of bounded variation, denoted as \(BV(\mathbb{R})\), and locally absolutely continuous functions, denoted as \(AC_{\mathrm{loc}}(\mathbb{R})\). The study considers generalized sampling series modified by a discrete kernel and an integrable kernel, paying special attention to averaged-type discrete kernels, such as central B-splines, averaged Fejér kernels, and averaged Bochner-Riesz kernels.\N\NThe main contributions of the paper can be summarized as follows:\N\N1-The authors first establish a regularizing property, demonstrating that the Durrmeyer sampling operators map essentially bounded functions into locally absolutely continuous functions.\N\N2-They prove a variation-detracting property, showing that if a function belongs to \(BV(\mathbb{R})\), its corresponding Durrmeyer sampling series is absolutely continuous, and its total variation is bounded by the product of the \(L^1\) norms of the kernels and the variation of the function.\N\N3-A crucial technical result is established by expressing the derivative of the Durrmeyer sampling operator applied to an absolutely continuous function as a finite linear combination of sampling Kantorovich operators applied to a convolution product involving the derivative of the function and a symmetric version of the kernel.\N\N4-Leveraging these tools, the authors achieve a full characterization of the space \(AC(\mathbb{R})\) in terms of the convergence in variation of the Durrmeyer sampling series.\N\N5-Finally, the paper provides a quantitative estimate for the rate of approximation. This error estimate is given in terms of a variation-based modulus of smoothness, which naturally reformulates the classical \(L^1\)-modulus of smoothness into the \(BV(\mathbb{R})\) setting.\N\N6-Furthermore, qualitative results for functions belonging to suitable Lipschitz classes are derived as a direct consequence.\N\NOverall, the manuscript provides a comprehensive functional analytic treatment of Durrmeyer sampling-type operators, highlighting their efficacy and regularizing properties as approximation tools in spaces of absolutely continuous functions on \(\mathbb{R}\).



Cites work









This page was built for publication: Durrmeyer sampling-type operators: approximation in variation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7233500)