Track Correlation Based on A´trous Algorithm

This paper mainly discusses the track correlation problem for the distributed multi-sensor multi-target tracking system. Target track data sequence can be seen as a signal in a certain period of time. In the point of signal processing, the signal can be divided into low-frequency and high-frequency two parts, low-frequency represents the overall trend of the signal, and high-frequency can be seen as the details characteristics of the signal and noise. In the wavelet analysis, they correspond to the scale coefficients and wavelet coefficients respectively. A´trous algorithm has the advantages of translation invariance, does not require sampling and interpolation etc. The details characteristics of the signal can be obtained more easily and storage capacity is reduced. It also has the advantages of less calculation time and easy programming. So the track sequences are decomposed to get the overall trend and local details of the track by use of a´trous algorithm. The simulation result shows the algorithm is more effective.