Absolute \(N_{q_\alpha}\)-summability of the series conjugate to a Fourier series (Q1304596)

From MaRDI portal
Revision as of 02:51, 5 March 2024 by Import240304020342 (talk | contribs) (Set profile property.)
scientific article
Language Label Description Also known as
English
Absolute \(N_{q_\alpha}\)-summability of the series conjugate to a Fourier series
scientific article

    Statements

    Absolute \(N_{q_\alpha}\)-summability of the series conjugate to a Fourier series (English)
    0 references
    0 references
    18 May 2000
    0 references
    Let \(q_\alpha(t)\) \((\alpha\geq 0)\) be the class of functions as defined in \textit{M. Samal} [J. Indian Math. Soc., New Ser. 50, No. 1-4, 191-209 (1986; Zbl 0688.42009)]. Then \(\sum_{n=0}^\infty a_n\) is said to be absolutely summable by \(N_{q_\alpha}\)-method or \(|N_{q_\alpha}|\)-summable if \[ \int_A^\infty w^{-2} \Biggl|\sum_{n\leq\omega} n a_n q_\alpha(n/w) \Biggr|dw<\infty, \] where \(A\) is a positive constant. Let \(f\in L_{2\pi}\) and let \(x\in \mathbb{R}\). Then define \[ \begin{alignedat}{2} \varphi(t) &=\tfrac 12 \{f(x+t)+ f(x-t)\}, \quad \psi(t)= \tfrac 12 \{f(x+t)- f(x-t)\},\\ \varphi_\alpha(t)&= \frac{\alpha}{t^\alpha} \int_0^t (t-u)^{\alpha-1} \varphi(u) du \quad (\alpha>0),&&{}\\ \Psi_\alpha(t)&= \frac{1}{\Gamma(\alpha)} \int_0^t (t-u)^{\alpha-1} \psi(u) du \quad (\alpha>0), \quad \psi_\alpha(t)= \Gamma(\alpha+1)t^{-\alpha} \Psi_\alpha(t). \end{alignedat} \] It was known to \textit{M. Samal} [Ph.D. Thesis in Mathematics, Utkal University India (1991)] that the Fourier series of \(f\) at a point \(x\) is \(|N,q_\alpha|\)-summable if \(\varphi_\alpha(t)\in \text{BV} (0,\pi)\), \(\alpha>0\). In the present paper, the author has proved its analogue for the conjugate series of the Fourier series. Theorem. Suppose \(\alpha>0\). If \(\psi_\alpha(0+)=0\) and \(\int_0^\pi t^{-\alpha}|\alpha\Psi_\alpha(t)|< \infty\), then the series conjugate to the Fourier series of \(f\) is \(|N_{q_\alpha}|\)-summable at a point \(x\). The author has also shown that if we let \(q_\alpha(t)= \beta(1-t)^{\beta-1}\), where \(0< \alpha< \beta< h+1\), \([\alpha]=h\), in the theorem above, then the method \(|N_{q_\alpha}|\) could be replaced by \(|C,\beta|\) \((\beta> \alpha> 0)\).
    0 references
    absolute summability
    0 references
    Fourier series
    0 references
    conjugate series
    0 references
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references