Wireless sensor networks (WSNs) are vulnerable to a variety of malicious attacks, especially the packet dropping attack, making security an important research field. Since prevention based techniques are less helpful for guarding against inside attacks, intrusion detection (ID) techniques are indispensable to provide advanced protection. This paper proposes an innovative cellular ID framework for packet dropping attack, which takes the deployment of passive listening nodes into consideration. Performance evaluation made in VisualSense demonstrates its high detection accuracy.