Research on the Real-time Registration Technique for Radar Networking
You He, Yun-long Dong, Guan Cheng-bin, Guohong Wang · 2006
Since the system errors degrade the association and fusion of the tracks from different radars greatly, registration is the vital problem for the data fusion of the radar network. But the measurements are always nonlinear function of the system biases; therefore, Kalman filter is unable to be used directly two methods are proposed in this paper to solve this problem. First, we use the linear model of literature (M.P Dana, 1990), and present an extended Kalman filter. Second, a sequential Monte Carlo approach is applied to real-time estimation of the state and the system errors, this method is known as particle filtering (M.Sanjeev Arumpalam et al., 2002) also. In the end, simulation results show the effectiveness of the two methods.