Hidden Markov Models (HMMs) are a common tool in pattern recognition. Applications of HMMs include voice recognition [1,2], texture recognition [3], handwriting recognition [4,5], gait recognition [6], tracking [7], and human behavior recognition [8,9]. Variations of these applications can also be used in distributed sensor networks. A lot of work has been done using HMMs to identify target behaviors from sensor data. For example, in Refs. [10,11], HMMs are used for activity monitoring and action recognition in body sensor networks. In Ref. [12], Amutha et al. used HMMs for target tracking in wireless sensor networks. HMMs can also be used to solve sensor network localization problem, which estimates the geographical location of sensors in wireless sensor networks [13]. Using HMMs, it is possible to design data transmission protocols in sensor networks for energy efficiency [14]. Other possible application for HMMs in sensor networks can be side channel analysis, such as power monitoring and network traffic timing analysis. This kind of analysis can extract extra source of information that can be exploited to analyze the sensor networks.