An energy-efficient approach for target recognition with synchronous sampling sensors

With rapid improvements in computer and software engineering, collaborative target recognition has become an active area of research in wireless sensor network community. Typically, each sensor node is designed to work using its very restrict battery capacity and limited computational resources, while the base station usually has sufficient energy supply and computational power. Hence, minimizing energy consumption of sensors while maintaining a given classification accuracy is a key issue in this research area. This paper proposes an energy-efficient approach for target recognition in synchronous sampling wireless sensor network, inspired by the autonomy of agents in the sensors. The experiments show that the energy consumption of the system can noticeably be reduced without losing much recognition rate.