Generalized Cullen numbers in linear recurrence sequences (Q2420396): Difference between revisions

Generalized Cullen numbers, are numbers written on the form \(C_{m,s}=ms^m+1\). In this paper, inspired by previous works, the authors investigate the existence of generalized Cullen numbers on linear recurrence sequences. More generally, given a linear recurrence sequence \(\{G_n\}\), \(a\in\mathbb{Z}\) and a polynomial with integer coefficients \(T(x)\), they study the solutions of the Diophantine equation \(G_n=ax^m+T(x)\) (note that if \(T(x)=1\), this implies finding generalized Cullen numbers among the elements of the sequence). The main result of the paper states that under certain assumptions over the roots of the characteristic polynomial of the sequence, there exist effectively computable constants \(C_1\) and \(C_2\) such that if \((m,n,x)\) is a solution of the equation, then \(m\leq C_1\) and \(n\leq C_2\log |x|\). Then, they proceed to compute \(C_1\) and \(C_2\) in the case of Cullen numbers (\(m=n\), \(s=2\), \(T(x)=1\)) and \(G_n\) the sequence of the so-called \(k\)-Fibonacci numbers recursively defined by \(F^{(k)}_n=\sum_{i=1}^kF^{(k)}_{n-i}\).