The Research On Performance Evaluation Of Filtering Localization Algorithm

Filtering localization algorithm is a kind of single observer passivelocalization algorithm, which makes use of the data of mixed with noises togauge the target state. As one of the hot spots of the passive location, filteringlocalization algorithm has an extremely important military significance of themodern information warfare for the advantages of simplicity in the facility,excellent invisibility, large effective radius and wide applicability. As thefiltering localization algorithm is emerging, It’s one of the most importantresearch topic for filter algorithm, that how to choose the optimal filteralgorithm based on the performance evaluation of filtering localizationalgorithm under specific conditions of use. Thus it is necessary and helpful toresearch on the performance evaluation of filtering localization algorithm, anddevelop an experiment platform with high openness and function expansion,which is convenient to load several kinds of filtering localization algorithmwhile doing an experiment, a simulation experiment or a performanceevaluationIn this paper, the comprehensive evaluation of filtering localizationalgorithm’s performance was researched and an evaluation platform wasdeveloped for filtering localization algorithm. Main content contains thefollowing four aspects:On the basis of detailing analysis of algorithm assessment methods, anevaluation indicator system was created for Kalman filtering algorithms, andan evaluation platform was built, then a case of EKF algorithm’s performanceevaluation using the analytic hierarchy process was gave. In the part offiltering localization algorithm, an infrared assisted filtering localizationalgorithm was created. Simulation results show that, the algorithm waseffective to improve the accuracy and convergence speed, compared withbear-only filtering localization algorithm. In the part of evaluation methodresearch, by simulation of nonlinear evaluation function, comparative study ofthe performance of the AHP comprehensive evaluation method andAHP-fuzzy comprehensive evaluation method, AHP-fuzzy comprehensiveevaluation method with better performance was choose as comprehensive performance evaluation methods for filtering localization algorithm. Finally,based on the study of filtering localization algorithm and assessment methods,an experiment evaluation platform was designed, then the validity andeffectiveness of the platform was verified by a special case.