Sensor networks have recently received an increasing interest both in industry and academia and are applicable to a wide range of distributed applications. In such networks, sensors organize themselves to collect or disseminate information in the network, routing messages to specific sensors, {\em etc}. Given the limited resources of those devices, energy consumption is a first class concern. At the core of data management in distributed systems, lie some basic functionalities such as broadcast or anycast. In this paper, we focus on providing a suite of *-cast (anycast, $k$-cast, broadcast) functionalities in a fully decentralized manner. More specifically, we present the design and evaluation of {\sc Solist}, implementing the *-cast suite for wireless sensor networks. {\sc Solist} is fully decentralized, and let the sensors get organized into a multi-layer structure, largely inspired from structured peer-to-peer systems, yet limiting the overall energy consumption. A type is associated to each sensor, and the \mbox{*-cast} functionalities are implemented at a type granularity regardless of the number of types and their distribution within the network. This enables to reach any (anycast), $k$ ($k$-cast), or all (broadcast) sensors of a given type in a large-scale network. Sensor nodes of the same type are logically clustered and each of those types remains reachable from any point in the system as long as the network is connected. We evaluate {\sc Solist} through simulation and show that {\sc Solist} provides those functionalities while achieving a reasonable trade-off between performance and energy consumption.