Random sampling compressive sensing (RSCS) is a prominent compression algorithm suitable for wireless sensor network as it can significantly reduce the number of samples required for signal reconstruction. This improves the network lifetime, but can introduce uncertainty in the signal reconstruction. In this paper, we propose a two-layer controller that ensures efficient signal reconstruction (i.e., low level of reconstruction error) and high network lifetime. The proposed controller is composed of a global controller, developed at the fusion center layer, and two local controllers, implemented at each node layer. The former steers the RSCS reconstruction error to a desired value, while the latter are implemented to reduce the energy consumption of each node. A performance evaluation of the proposed scheme is carried out in terms of reconstruction error regulation and network lifetime. Simulation results show the effectiveness of the proposed scheme compared with two main strategies existing in the literature.