The document proposes a new probabilistic coverage protocol (PCP) for sensor networks that can employ different sensing models. PCP aims to address the costly task of designing and testing different coverage protocols for each sensing model. It works with common disk sensing models as well as probabilistic sensing models with minimal changes. Simulation results show that PCP outperforms other deterministic and probabilistic protocols in terms of number of activated sensors, total energy consumed, and network lifetime while being robust against failures and inaccuracies.