Time Synchronization Algorithm in Sensor Network Based on Kalman Filter

Time information is critical for a variety of applications in sensor network. This work proposes a novel time synchronization algorithm based on Kalman filtering between all sensor nodes and a sink node. In our scheme, sink node periodically send beacon messages to sensor nodes all around using the network´s physical-layer broadcast. The recipients recode arrival time of the message by it own clock and estimate the frequency of node clock based on Kalman Filter algorithm. We also describe an experiment that characterizes its performance high precision synchronization using very little energy.