Notes on Fourier series: Strong approximation (Q762705): Difference between revisions

The paper is closely related to an earlier paper of the author [Functions, Series, Operators. Proc. Int. Conf. Budapest 1980, Vol. II, Colloq. Math. Soc. János Bolyai 35, 1195-1199 (1983; Zbl 0542.42005)] and it extends some results in the theory of so called strong approximation of Fourier series. Denote by \(C_{2\pi}\) the class of all continuous functions \(f: R\to R\) that are periodic with the period \(2\pi\). Let \(E_ n(f)\) stand for the error in the best uniform approximation of \(f\in C_{2\pi}\) by trigonometric polynomials. By \(s_ k=s_ k(f)\) \((k=1,2,...)\) we denote the partial sums of the Fourier series of \(f\in C_{2\pi}\). Let \(\Phi\) : [0,\(+\infty)\to [0,+\infty)\) be a continuous increasing function with \(\Phi (0)=0\). Extending an earlier result of \textit{G. Alexits} and \textit{D. Králik} [Publ. Math. Inst. Hungar. Acad. Sci., Ser. A 8 (1963), 317-327 (1964; Zbl 0129.050)] the author proves that the inequalities \((1/n)\sum^{2n}_{k=n+1}\Phi (| s_ k(x)-f(x)|)\leq K\Phi (E_ n(f))\) (for \(x\in [0,2\pi])\) hold for each \(f\in C_{2\pi}\) if and only if there exists an \(A>0\) such that \(\Phi (t)\leq e^{A.t}\) \((t\in [0,+\infty))\) and \(\Phi\) (2t)\(\leq A\Phi (t)\) (t\(\in (0,1))\). The author investigates also this question: What smoothness properties for the function f can be derived from the inequality \(\sup_{0\leq x\leq \pi}\sum^{\infty}_{k=0}\Phi (| s_ k(x)-f(x)|)<+\infty\).