Writing Numbers in Base 3, the Hard Way

From MaRDI portal





Let \(n\) be a natural number \(> 1\). A finite set \(A\) of integers is called a coefficient system in base \(n\) if every integer \(z\) can be written in the form \(z = a_0 + a_1 \cdot n + a_2 \cdot n^2 + \cdots + a_k \cdot n^k\) for some integer \(k \geq 0\) with each coefficient \(a_i\) in \(A\), and a coefficient system \(A\) in base \(n\) is called a number system in base \(n\) if there are exactly \(n\) numbers in \(A\). In [Publ. Math. 58, 549-557 (2001; Zbl 0980.11008)] the author proved the following theorem: If \(B = \{b_0 = 0,b_1,\ldots,b_{n-1}\}\) is a set of representatives of the congruence classes modulo \(n\) then the set \[ B - B = \{b_i - b_j \mid 0 \leq i,j \leq n-1\} \] is a coefficient system in base \(n\) if and only if \(\gcd(b_1,b_2,\ldots,b_{n-1}) = 1\). (In this case \(B\) is called a just touching covering system.) Here the author characterizes in an elementary way those sets \(\{0,a,b\}\) which are number systems in base \(3\) and derives some necessary conditions for a set \(\{0,a,b\}\) to be a number system in base \(3\).











This page was built for publication: Writing Numbers in Base 3, the Hard Way

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2773160)