On dimension of porous measures (Q697375)

From MaRDI portal
scientific article
Language Label Description Also known as
English
On dimension of porous measures
scientific article

    Statements

    On dimension of porous measures (English)
    0 references
    0 references
    17 September 2002
    0 references
    Porosity of a set is a quantity that describes the relative size of holes in the set at all small scales. It is well known that if the porosity of a set \(A\subset\mathbb R^n\) is close to its maximum value, then the dimension of \(A\) cannot be much bigger that \(n-1\). This was first proved for Hausdorff dimension by \textit{P. Mattila} [J. Lond. Math. Soc., II. Ser. 38, No. 1, 125--132 (1988; Zbl 0618.28005)] and later generalized for packing dimension with the correct asymptotics in terms of large porosity by \textit{A. Salli} [Proc. Lond. Math. Soc., III. Ser. 62, No. 2, 353--372 (1991; Zbl 0716.28006)]. These results were extended to mean porous sets having holes of certain size at a certain percentage of small scales by \textit{P. Koskela} and \textit{S. Rohde} [Math. Ann. 309, 593--609 (1997; Zbl 0890.30013)]. They found the correct asymptotics for the dimension estimate of such sets when the holes are small. In the paper under review the authors study corresponding questions for measures. They extend the earlier results by \textit{J.-P. Eckmann}, \textit{E. Järvenpää} and \textit{M. Järvenpää} [Nonlinearity 13, 1--18 (2000; Zbl 1022.28002)] concerning dimensional behaviour of porous measures satisfying the doubling condition by considering general measures and mean porosity. The results of the paper may be regarded as analogues of the results of Salli and Koskela-Rohde for measures. The proofs combine new ideas to the methods of Salli, Koskela-Rohde and Eckmann-Järvenpää-Järvenpää in an interesting way. An important tool of independent interest is to verify that for a mean porous measure there exists a mean porous set of positive measure with asymptotically the same percentage of porous scales as the measure itself.
    0 references
    0 references
    mean porous measure
    0 references
    porous measure
    0 references
    dimension
    0 references
    0 references
    0 references
    0 references