Study Of Novel Single And Multi Observer Passive Radar Localization Method Based On Computational Intelligence

Passive localization technology based on external illuminator has an importantresearch value. This paper firstly analyzes the characteristic of commonly usedlocalization method and then introduces a single observation localization model usingTOA measurements based on multi external illuminators. After that, a detailedmathematical analysis about the localization performance is introduced. Finally twoaccuracy analysis experiments about the influence of distance measuring error andilluminators site distribution are given.The essence of target localization is to solve complex nonlinear problems.Traditional solutions have some apparent defects because they are often sensitive toinitial value, converge too slow and require linearization process, which result in thecondition of low localization accuracy and restrict the application of passive localizationtechnology in actual engineering. According to the favorable performance ofcomputational intelligence technology on solving engineering optimization problems,this paper respectively presents a novel multi observation passive localization methodbased on Memetic algorithm and a novel single observation passive localization methodbased on MOEA. Moreover, a novel Memetic local search strategy and a specificdesigned multi-objective decision making strategy are also proposed. Finally, a group ofsimulation experiments are given to evaluate the performance of new algorithm fromdifferent perspective.Experimental results show that the presented algorithms, which approachCramer-Rao lower Bound excessively can not only operate steadily and convergerapidly but also can avoid the defects of traditional methods. So a conclusion can bedrawn that solving the nonlinear optimization problem of passive localization systemwith computational intelligence technology is available and feasible.