Circular average relative to fractal measures (Q2079195)

The paper under review is concerned with local smoothing estimates for the wave equation associated with fractal measures and their consequences for the spherical averaging operator. Let \(f\) be a function whose Fourier transform is supported on the annulus \(\{\xi\in \mathbb{R}^{d} : \lambda \leq |\xi| \leq 2\lambda\}\) for \(\lambda \geq 1\). For the half-wave operator \[ Wf(x,t) = \int e^{i(x\cdot \xi + t|\xi|)} \widehat{f}(\xi) d\xi, \] the paper studies local smoothing estimates of the form \[ \| W f \|_{L^q(\mathbb{R}^d\times [1,2], \nu)} \leq C \lambda^\gamma \| f\|_{L^p(\mathbb{R}^d)}, \] where the constant \(C\) is independent of \(f, \lambda\) and \(\nu\), where \(\nu\) is a non-negative Borel measure on \(\mathbb{R}^{d}\times (0,\infty)\) such that \[ \nu(B_r) \leq r^{\alpha} \] for some \(0 < \alpha \leq d+1\) for every ball \(B_r \subset \mathbb{R}^{d+1}\) of radius \(r\). The main question is the following: For given \(p\), \(q\), \(\alpha\), what is the optimal power \(\gamma\) for which the estimate holds? Even when \(\nu\) is the Lebesgue measure, this is an important open problem for \(d\geq 3\). This article focuses on the case \(d=2\), where the problem is settled (up to \(\epsilon\)) for the Lebesgue measure by \textit{L. Guth} et al. [Ann. Math. (2) 192, No. 2, 551--581 (2020; Zbl 1450.35156)]. For \(d=2\), based on examples, the authors conjecture that the above estimate holds for \(\gamma > \frac{1}{2}+\frac{1}{p}-\frac{\alpha}{q}\) whenever \(p\leq q\) and \(\frac{1}{p}+\frac{\kappa(\alpha)}{q} \leq 1\) for a certain index \(\kappa(\alpha)\) and prove it for partial ranges of \(p,q\). Via the well-known connection between spherical averages and the wave equation, sharp \(L^p\) improving estimates for the circular maximal function relative to fractal measures are obtained in a certain \(p\)-range. The proof uses techniques related to the Fourier restriction problem. Main ingredients include Bennett-Carbery-Tao multilinear restriction estimates, Bourgain-Guth induction on scales argument adapted by Lee-Vargas, and sharp local smoothing estimates due to Guth-Wang-Zhang.