The initialization problem, also known as naming, consists to give a unique identifier ranging from $1$ to $n$ to a set of $n$ indistinguishable nodes in a given network. We consider a network where $n$ nodes (processors) are randomly deployed in a square (resp. cube) $X$. We assume that the time is slotted and the network is synchronous, two nodes are able to communicate if they are within distance at most of $r$ of each other ($r$ is the transmitting/receiving range). Moreover, if two or more neighbors of a processor $u$ transmit concurrently at the same time slot, then $u$ would not receive either messages. We suppose also that the anonymous nodes know neither the topology of the network nor the number of nodes in the network. Under this extremal scenario, we first show how the transmitting range of the deployed processors affects the typical characteristics of the network. Then, by allowing the nodes to transmit at various ranges we design sub-linear randomized initialization protocols~: In the two, resp. three, dimensional case, our randomized initialization algorithms run in $O(n^{1/2} \log{n}^{1/2})$, resp. $O(n^{1/3} \log{n}^{2/3})$, time slots. These latter protocols are based upon an optimal gossiping algorithm which is of independent interest.
