Uniform convergence and integrability of Fourier integrals (Q2637951)

Let \[ F(x)=\int_{0}^\infty f(t)\cos xt\,dt \] denote the cosine transform and similarly \[ G(x)=\int_{0}^\infty g(t)\sin xt\, dt \] denote the sine transform when the integrals exist. Also, \(S_N(F)\) denote the cosine transform when the integration is taken over \([0,N]\), with a similar definition for \(S_N(G)\). In general, the functions of interest are defined on \(\mathbb{R}_+\), locally of bounded variation on \((0,\infty)\), vanishing at infinity and such that \(f\in L^1[0,1]\) and \(tg\in L^1[0,1]\). The expression \(\int_{0}^\infty|df(t)|\,dt\) denotes the bounded variation of the function \(f\). The pair of functions \((h,\beta)\) determines a general monotone function \(h\) with majorant \(\beta\) if for all \(x\in (0,\infty)\) \[ \int_{x}^{2x}|dh(t)|\,dt< C\beta(x). \] When this happens, write \((h,\beta)\in GM\). Among the numerous results in the paper, the authors are able to show that the following estimates hold: \[ \begin{aligned} \left\| F-S_N(F)\right\|_{\infty} & \leq \max_{x>N}\left|\int_{N}^x f(t)\,dt \right|+ C\max_{x\geq\frac{N}{2}} \left(x\int_{x}^{2x}|df(t)|\right)\\ \left\| G-S_N(G)\right\|_{\infty} & \leq C \max_{x\geq\frac{N}{2}}\left(x\int_{x}^{2x}|dg(t)|\right). \end{aligned} \] Using these estimates, the authors are able to obtain uniform convergence results for the sine and cosine integrals for general monotone functions. In particular, an example of one of their results is the following: Theorem: Suppose that \((g,\beta)\in GM\) where (for some \(c>1\)) \[ \beta(t)=\frac{1}{t}\max_{u\geq\frac{t}{c}}\int_{u}^{2u}g(s)\,ds. \] Then \(G\) converges uniformly on \(\mathbb{R}_+\) if and only if \(\lim_{t\to\infty} tg(t)=0\). Additionally, the authors use the estimates to prove similar results for \(L^p\) integrability of the Fourier transform when integrated against power weights and results for the uniform convergence of the Fourier transform. The results are also extended to slightly different general monotone weight functions.