Track Correlation Algorithm Based on Discrete Wavelet Transform

A novel track-to-track association algorithm based on multiple 1-D discrete wavelet transform(DWT)is proposed in this paper. Since the target track data sequence is actually a non-stationary random process, we utilize discrete wavelet transform to analyze and process it. The ratio of energy in high resolution wavelet space to energy in low resolution scale space is introduced to establish hypothesis test model based on F-distribution. Combining it with decorrelation method we discussed the multi-dimension problem and represented a test method. Computer simulation results show that the algorithm is effective and has a good performance to distinguish two cross tracks.