Blind estimation of number of motion multi-human targets in wireless pyroelectric infrared sensor networks

• We present a method which estimates number of motion human in sensor network.
• The independent criterion is modeled by Bayesian network.
• The objective function is optimally solved by the Laplace estimation.
• Numerous experiments are completed by using LiTaO3 pyroelectric infrared sensor.

Due to the features of low energy consumption and flexible networking, nowadays the pyroelectric sensor has been applied widely in areas such as network instruction detection or human body target tracking recognition. Moreover, accurate estimation and judgment about the number of human targets moving in the networks is the foundation of tracking and recognition. This paper, under the condition of being lack of relevant prior knowledge, presents a novel method which selects the maximum likelihood function of the Bayesian network models as the independent criterion. In addition, the objective function is optimally solved by the Laplace estimation. The results of numerous experiments on both simulation and hardware experimental platforms are shown that this method has capability to blindly estimate the number of motion multiple human targets in wireless pyroelectric infrared sensor networks.