The invention relates to a precise
clock synchronization parameter tracking method based on an augmented Kalman neural network, and belongs to the field of
time synchronization. According to the method, for a multi-hop network scene with environment temperature change and asymmetric link
transmission delay, the influence of the accumulated asymmetric link
transmission delay in a multi-hop
network on clock synchronization precision is considered, and a multi-hop
network clock observation equation containing the accumulated asymmetric link
delay is obtained; establishing a recursive state equation of the
clock frequency offset and the temperature change based on a quadratic polynomial model of the temperature and the clock
frequency offset; and modeling an evolution process of accumulated asymmetric
delay and an unknown
temperature coefficient into a first-order linear difference equation, augmenting the first-order linear difference equation to a clock parameter tracking
state space model, and realizing joint tracking of
clock phase offset and clock
frequency offset of a slave node by using an augmented Kalman neural network
algorithm. According to the method, joint tracking of
clock phase offset and frequency offset is carried out by adopting the idea of data and model joint driving, and the accuracy and robustness of clock parameter tracking are improved.