Spectral property of self-affine measures on \(\mathbb{R}^n\) (Q5891018)

A Borel probability measure \(\mu\) on \(\mathbb R^n\) is said to be a spectral measure if there is a discrete set \(\Lambda\subset\mathbb R^n\) for which the set of complex exponentials \(\{\exp(2\pi\mathrm i\langle\lambda,x\rangle)\mid\lambda\in\Lambda\}\) forms an orthonormal basis of \(L^2(\mathbb R^n,\mu)\); one of the motivations for their study is a conjecture of \textit{B. Fuglede} [J. Funct. Anal. 16, 101--121 (1974; Zbl 0279.47014)] that the indicator function of a set is the density with respect to Lebesgue measure of a spectral measure if and only if the set is a translational tile (the conjecture remains open for \(n\leq 2\) and is known to be false for \(n\geq 3\)). Here a measure \(\mu_{M,D}\) is constructed from an expanding \(n\times n\) integer matrix \(M\) and digit set \(D\) as the unique self-affine invariant probability measure associated to the iterated function system given by the contractive maps \(x\mapsto M^{-1}(x+d)\) for \(d\in D\). Sufficient conditions for \(\mu_{M,D}\) to be a spectral measure or to have infinitely many orthogonal exponentials are found and in special cases necessary and sufficient conditions are found for spectrality. These results extend work of \textit{X.-R. Dai} et al. [Adv. Math. 242, 187--208 (2013; Zbl 1277.28009); Adv. Math. 259, 511--531 (2014; Zbl 1303.28011)] to higher dimensions.