Quenched limits for the fluctuations of transient random walks in random environment on \(\mathbb{Z}\) (Q1950266): Difference between revisions

This paper deals with nearest-neighbour random walks in random environment (RWRE) on \(\mathbb{Z}\), a model introduced by \textit{F. Solomon} [Ann. Probab. 3, 1--31 (1975; Zbl 0305.60029)]. The random environment is a sequence of independent and identically distributed random variables taking values in \((0,1)\), indexed by \(\mathbb{Z}\); at each step the walk goes to the left or to the right with probability depending on the value of the environment at its current location. The authors consider transient RWRE under the quenched law, that is for a typical and fixed environment, which is relevant for applications like DNA unzipping. They are interested in the fluctuations of the hitting time of a level \(x\) around its (quenched) mean. They prove that for large \(x\) and for a set of environments arbitrarily close to 1, these fluctuations can be described explicitly by a function of the environment, and that their limiting law can be characterized by means of a Poisson point process with explicit intensity. Therefore, their result can be seen as the quenched counterpart of the one by \textit{H. Kesten, M. V. Kozlov} and \textit{F. Spitzer} [Compos. Math. 30, 145--168 (1975; Zbl 0388.60069)]. The proof relies on a decomposition of the so-called potential into valleys. The fluctuations of the hitting times are then related to the crossing times of valleys, the main contribution coming from deep valleys, which are shown to be asymptotically independent. To improve readability, the result is followed by a sketch of the proof and preparatory lemmas, while some of the most technical parts are deferred to an appendix. This paper will be of interest for anyone working in the field of random walks in random environment.