Comparison of the KF and particle filter based out-of-sequence measurement filtering algorithms

Abstract

Current out-ofsequence measurement (OOSW filtering algorithms belong to two distinct classes, Kalman filter (U) or extended KF (EKF) based and particle filter (PF) based. This paper compares the performances of the multiple-lag I@ and PF based OOSMJiltering algorithms for a number of scenarios with linear dynamic and measurement models with additive Gaussian noisesjrst. The KF with in-sequence measurements represents an optimal estimator. Therefore, for this case, we compare the performances of the OOSMjltering algorithms relative to the KF with in-sequence measurements. Numerical results show that the KF OOSMaIgorithm used in this study is optimal. Next, we evaluate the multiple-lag KF/EU and PF based 0OSMf;rtering algorithms using realistic Ground Moving Target Indicator (GMTI) sensor measurements. We use estimation accuracy and statistical consistency for comparison.