How many squares must a binary sequence contain?
We consider numeration systems with base β and − β, for quadratic Pisot numbers β and focus on comparing the combinatorial structure of the sets Zβ and Z− β of numbers with integer expansion in base β, resp. − β. Our main result is the comparison of languages of infinite words uβ and u− β coding the ordering of distances between consecutive β- and (− β)-integers. It turns out that for a class of roots β of x2 − mx − m, the languages coincide, while for other quadratic Pisot numbers the language...
Let be a non-integer. We consider expansions of the form , where the digits are generated by means of a Borel map defined on . We show existence and uniqueness of a -invariant probability measure, absolutely continuous with respect to , where is the Bernoulli measure on with parameter () and is the normalized Lebesgue measure on . Furthermore, this measure is of the form , where is equivalent to . We prove that the measure of maximal entropy and are mutually singular. In...
Let be a non-integer. We consider -expansions of the form , where the digits are generated by means of a Borel map defined on . We show that has a unique mixing measure of maximal entropy with marginal measure an infinite convolution of Bernoulli measures. Furthermore, under the measure the digits form a uniform Bernoulli process. In case 1 has a finite greedy expansion with positive coefficients, the measure of maximal entropy is Markov. We also discuss the uniqueness of -expansions....
We consider minimal redundant digit expansions in canonical number systems in the gaussian integers. In contrast to the case of rational integers, where the knowledge of the two least significant digits in the “standard” expansion suffices to calculate the least significant digit in a minimal redundant expansion, such a property does not hold in the gaussian numbers : We prove that there exist pairs of numbers whose non-redundant expansions agree arbitrarily well but which have different least significant...
Let E be an interval in the unit interval [0,1). For each x ∈ [0,1) define dₙ(x) ∈ 0,1 by , where t is the fractional part of t. Then x is called a normal number mod 2 with respect to E if converges to 1/2. It is shown that for any interval E ≠(1/6, 5/6) a.e. x is a normal number mod 2 with respect to E. For E = (1/6, 5/6) it is proved that converges a.e. and the limit equals 1/3 or 2/3 depending on x.