A method for the numerical resolution of Abel-type integral equations of the first kind (Q1109547)

From MaRDI portal
scientific article
Language Label Description Also known as
English
A method for the numerical resolution of Abel-type integral equations of the first kind
scientific article

    Statements

    A method for the numerical resolution of Abel-type integral equations of the first kind (English)
    0 references
    1988
    0 references
    The integral equation \[ \int^{x}_{0}(x-t)^{- \alpha}k(x,t)y(t)dt=g(x),\quad 0\leq x\leq 1,\quad 0<\alpha <1 \] is considered under appropriate continuity and smoothness assumptions on the functions \(k\) and \(q\), in particular \(k(x,x)\neq 0\). The interval \([0,1]\) is equdistantly partitioned, and in each subinterval a two-point sinusoidal interpolation ansatz is used in collocative manner. In analogy to \textit{P.P.B. Eggermont}'s analysis of the trapezoidal discretization method [Numer. Math. 52, 65--79 (1988; Zbl 0634.65138)] the author proves that convergence of the approximate solution is of order 2 in the maximum norm, provided a particular condition by numerical calculation for \(0.10321 \leq \alpha < 1\). Finally, she displays results of a few numerical case studies.
    0 references
    0 references
    Abel integral equations
    0 references
    product quadrature rules
    0 references
    trigonometric interpolation
    0 references
    convergence.
    0 references
    0 references