Wireless sensor networks (WSNs) have been widely studied and usefully employed in many applications such as monitoring environment, embedded system and so on. In WSNs, substantial nodes are deployed randomly over the entire desired area; therefore, the sensing regions of different nodes may be partially overlapped. This is referred to as the sensing coverage problem. In this paper we first define a maximum sensing coverage region problem (MSCR) in WSNs and solve the problem by the proposed algorithm. In our method, the maximum monitored area fully covered by the minimum active sensors. We apply the proposed algorithm to improve ARPEES, an adaptive routing protocol for WSNs. By integrating MSCR and ARPEES, we propose a new architecture for routing protocol in large scale distributed WSNs. With the distributed and light overhead traffic approach, we aim to balance energy consumption required for sensing data, forming cluster, selecting cluster head, and relaying data to different sensor nodes to prolong the whole network lifetime. The simulation program was developed to evaluate the performance of the protocol and compare with other existing routing protocols.