An energy-efficient sub-Nyquist sampling method based on compressed sensing in wireless sensor network for vehicle detection

Abstract

The wireless magnetic sensor network is scalable and deployable for traffic surveillance. But active magnetic sensors of the wireless sensor node have high energy consumption which cannot be ignored. It is necessary to save energy of the wireless magnetic sensor node for vehicle detection. In this paper, based on compressed sensing (CS) by random down sampling matrix, an energy-efficient sub-Nyquist sampling method in magnetic sensor network is proposed for vehicle detection. With this new sampling method, the active magnetic sensor's average frequency is less than the Nyquist standard sampling frequency, which reduces the energy consumption of the active sensor, while extending the lifetime of the wireless sensor nodes. When the Compressed Radio (CR) meets the maximum value of 60%, the new sampling method doubles the wireless magnetic sensor node's lifetime and maintains vehicle detection accuracy.