1: \begin{abstract}

2:   A binary labeling of a graph is a function from its nodes to

3:   $\{0,1\}$.  Scenery reconstruction is the problem of inferring a

4:   labeling given the labels observed by a particle performing a random

5:   walk on the graph.

6: 

7:   We consider the question of when a random walk on a finite abelian

8:   group with a given step distribution is {\em reconstructive}, or can

9:   be used to reconstruct a binary labeling up to a shift. We focus

10:   initially on walks on undirected cycles. Matzinger and Lember (2006)

11:   give a sufficient condition for reconstructibility on cycles,

12:   involving the Fourier transform of the random walk's step

13:   distribution. While, as we show, this condition is not in general

14:   necessary, our main result is that it is necessary when the length

15:   of the cycle is prime and larger than 5, and the step distribution

16:   has only rational probabilities.

17: 

18:   We use this result to show that rational random walks on prime

19:   cycles which have non-zero drift (for an appropriate notion of

20:   drift) are reconstructive, as is any random walk with a

21:   non-symmetric bounded step function, on a large enough cycle.

22:   

23:   We extend our results to walks on a large class of abelian groups,

24:   namely products of prime cycles. These include some cycles of

25:   composite length as well as regular tori of prime length.

26: \end{abstract}

27: