Parameter Estimation for Maneuvering Target in OTHR Relying on Improved Maximum-Likelihood Algorithm

摘要

This paper presents an improved MaximumLikelihood (ML) estimation method for maneuvering target parameters of over-the-horizon radar (OTHR). To avoid the matrix inversion involved in traditional ML function, the ML problem is reduced to 'over-determined' non-linear least squares problem. Genetic algorithm is used to estimate maneuvering target parameters with high accuracy under low signal-to-noise ratio (SNR). In addition, the Cramer-Rao Bound (CRB) for parameter estimation in OTHR is derived. Compared with the existing methods, the proposed algorithm has the following advantages: (1) higher estimation accuracy; (2) lower input SNR; (3) simultaneous estimation of parameters of multiple maneuvering targets. The simulation results show the superiority of the algorithm.