On the cyclicity of the rational points group of abelian varieties over finite fields

Abstract

We propose a simple criterion to know if an abelian variety $A$ defined over a finite field ${\mathbb{F}}_q$ is cyclic, i.e., it has a cyclic group of rational points; this criterion is based on the endomorphism ring End${}_{{\mathbb{F}}_q}(A)$. We also provide a criterion to know if an isogeny class is cyclic, i.e., all its varieties are cyclic; this criterion is based on the characteristic polynomial of the isogeny class. We find some asymptotic lower bounds on the fraction of cyclic ${\mathbb{F}}_q$-isogeny classes among certain families of them, when $q$ tends to infinity. Some of these bounds require an additional hypothesis. In the case of surfaces, we prove that this hypothesis is achieved and, over all ${\mathbb{F}}_q$-isogeny classes with endomorphism algebra being a field and where $q$ is an even power of a prime, we prove that the one with maximal number of rational points is cyclic and ordinary.

13 pages, this is a preliminary version, comments are welcome