The main role of wireless sensor networks (WSN) is to collect sensor environmental data from monitored area. Due to faults or malicious nodes, sensor data collected or reported might be wrong. Hence it is important to detect the presence of wrong sensor readings and misleading reports [18]. In this paper, we present a formal model using Time Petri Net to formally evaluate a proposed solution for detecting wormhole attack in CL-MAC, a cross-layer MAC protocol developed at MAC layer for energy efficient and low latency in WSN [3] [7]. Lunching wormhole attack in CL-MAC can cause the protocol invalid in some scenarios. Wormhole attack in CL-MAC is discussed, and a formal approach based on Time Petri Net (TPN) to detect wormhole attack is presented. TiNA (Time Net Analyzer) TPN tool is used to highlight the formal properties of the proposed solution. The obtained analytical results show that the secured version of CL-MAC can effectively detect and avoid wormhole attack, and make more sensor nodes perform a valid behavior.