错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Continuous Functions with Locally Complicated and Fractal Properties Related to Infinite-Symbol B-Representation of Numbers

  • Mykola Pratsiovytyi,
  • Olga Bondarenko,
  • Iryna Lysenko,
  • Sofiya Ratushniak

摘要

We introduce and study a massive class of continuous functions defined in the interval (0; 1) by using a special encoding (representation) of the argument with an alphabet Z = {0, ±1, ±2, . . .}: \(\begin{array}{c}x={b}_{{\alpha }_{1}}+\sum\limits_{k=2}^{m}{b}_{{\alpha }_{k}}\prod\limits_{i=1}^{k-1}{\Theta }_{{\alpha }_{i}}\equiv {\Delta }_{{\alpha }_{1}{\alpha }_{2}\dots {\alpha }_{m}\left(\mathrm{\varnothing }\right)}^{B},\\ x={b}_{{\alpha }_{1}}+\sum\limits_{k=2}^{\infty }{b}_{{\alpha }_{k}}\prod\limits_{i=1}^{k-1}{\Theta }_{{\alpha }_{i}}\equiv {\Delta }_{{\alpha }_{1}{\alpha }_{2}\dots {\alpha }_{n}\dots }^{B},\end{array}\) x = b α 1 + k = 2 m b α k i = 1 k - 1 Θ α i Δ α 1 α 2 α m B , x = b α 1 + k = 2 b α k i = 1 k - 1 Θ α i Δ α 1 α 2 α n B ,

where αn ∈ Z‚ Θn > 0 ∀n ∈ Z; \({\sum }_{n=-\infty }^{+\infty }{\Theta }_{n}=1\) n = - + Θ n = 1 , and \({b}_{n+1}\equiv {\sum }_{i=-\infty }^{n-1}={b}_{n}+{\Theta }_{n}\forall n\in Z\) b n + 1 i = - n - 1 = b n + Θ n n Z .

A function f; which is the main object of our investigations, is defined by the following equalities: \(\left\{\begin{array}{l}f\left(x={\Delta }_{{i}_{1}\dots {i}_{k}\dots }^{B}\right)={\sigma }_{{i}_{1}1}+{\sum }_{k=2}^{\infty }{\sigma }_{{i}_{k}k}{\prod }_{j=1}^{k-1}{p}_{{i}_{j}j}\equiv {\Delta }_{{i}_{1}\dots {i}_{k}\dots },\\ f\left(x={\Delta }_{{i}_{1}\dots {i}_{m}\left(\mathrm{\varnothing }\right)}^{B}\right)={\sigma }_{{i}_{1}1}+{\sum }_{k=2}^{m}{\sigma }_{{i}_{k}k}{\prod }_{j=1}^{k-1}{p}_{{i}_{j}j}\equiv {\Delta }_{{i}_{1}\dots {i}_{m}\left(\mathrm{\varnothing }\right)},\end{array}\right.\) f x = Δ i 1 i k B = σ i 1 1 + k = 2 σ i k k j = 1 k - 1 p i j j Δ i 1 i k , f x = Δ i 1 i m B = σ i 1 1 + k = 2 m σ i k k j = 1 k - 1 p i j j Δ i 1 i m ,

where the infinite matrix ||pik||, iZ; kN, satisfies the conditions \(\begin{array}{l}\left.1\right) \left|{p}_{ik}\right|<1 \forall i\in Z, \forall k\in N;\\ \left.2\right) {\sum }_{i\in Z}{p}_{ik}=1\forall k\in N;\\ \left.3\right) 0<{\sum }_{k=2}^{\infty }{\prod }_{j=1}^{k-1}{p}_{{i}_{j}j}<\infty \forall \left({i}_{j}\right)\in L;\\ \left.4\right) 0<{\sigma }_{ik}\equiv {\sum }_{j=-\infty }^{i-1}{p}_{jk}<1\forall i\in Z,\forall k\in N.\end{array}\) 1 p ik < 1 i Z , k N ; 2 i Z p ik = 1 k N ; 3 0 < k = 2 j = 1 k - 1 p i j j < i j L ; 4 0 < σ ik j = - i - 1 p jk < 1 i Z , k N .

The introduced class of functions contains monotone, nonmonotone, and nowhere monotone functions, functions without monotonicity intervals except the constancy intervals, Cantor-type and quasi-Cantor-type functions, and functions of bounded and unbounded variation. We establish criteria for the function f to be a monotonic and Cantor-type function, as well as a criterion of nowhere monotonicity of this function. We obtain expressions for the Lebesgue measure of the set of nonconstancy of this function and for the variation of the function. The necessary and sufficient conditions under which the function has unbounded variation are established. For a particular case, we describe the self-similarity (structural fractality) of the plot of the analyzed function and study its differential properties.