Data registration of radars based on RBUKF and performance analysis

This paper mainly discussed the problem of data registration algorithm of radars and the biased estimation with UKF (unscented Kalman filter). For the efficiency of UKF, the Rao-Blackwellised way is utilized for that registration model includes both linear and nonlinear parts. The RBUKF (Rao-Blackwellised unscented Kalman filter) is discussed with reduced unscented transform and tractable scheme in detail. Then a suitable RBUKF registration algorithm is developed and the theoretical Cramer-Rao lower bound (CRLB) of estimation error are discussed. Theoretical analysis of mean square error (MSE) are implement to evaluate performance of algorithm based on CRLB. Simulation shows RBUKF registration algorithm works efficiently and closely reach the CRLB.