Improve the Tracking Performance of Maneuvering Target Based on Wavelet Neural Network

Wavelet neural network (WNN) takes nonlinear wavelet bases as hidden nodes activation to replace nonlinear activation function in neural networks. It has the advantages of self-learning, rapid convergence rate and nonlinear approximation ability. Aiming at the maneuvering frequency is traditionally determined beforehand as a constant based on the target state estimation in the state model of the maneuvering target. An improved maneuvering target tracking method based on WNN is proposed. The input of the WNN is the new residual, the output of WNN is used to update the maneuvering frequency to realize the adaptive adjustment of the maneuvering frequency of the CS (current statistical) model. The improved algorithm is more close to the real state of the target. The simulation results showed that tracking error can be reduced and the tracking performance can be improved.