Answers are grounded in this paper and cite their sources.
Let N >= 2, 0 < ρ< 1, and let b_0, ..., bN-1 be distinct vectors in R^d. Let μbe the uniform homogeneous self-similar probability measure generated by the maps S_a(x) = ρx + b_a, without any separation assumption. We prove that if μadmits a frame measure, then the conditional coding measure is a Dirac mass for μ-almost every point. Equivalently, the coding map restricts to a Borel bijection between full-measure subsets of the symbolic space and the attractor. Consequently, the fiber entropy vanishes at every finite level, and μis exact dimensional with H μ= log N/-log ρ ≤ d. In particular, μadmits no frame measure whenever its Hausdorff dimension is strictly smaller than the similarity dimension. This excludes frame measures whenever Nρ^d > 1. As an application, we show that the uniform consecutive-digit Bernoulli convolution on the real line admits a frame measure in the range N⁻¹ <= ρ< 1 if and only if ρ= N⁻¹.
No takes yet. Share an insight, caveat, or question.
Wei-Jie Wang (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: