Variable Order Fractional Kalman Filters for Estimation over Lossy Network

Abstract

This paper presents generalization of Fractional Variable Order Kalman Filter (FvoKF) and Improved Fractional Variable Order Kalman Filter (ExFvoKF) algorithms for estimation of fractional variable order state-space systems over lossy network. This generalization is obtained for a state-space system based on one type of fractional variable order difference (\(\mathcal{A}\)-type) and assuming the knowledge about packets losing. The generalization of ExFKF algorithm based on the infinite dimensional form of a linear discrete fractional variable order state-space system and measurements equation augmentation. It required less restrictive assumptions than derivation of FKF. Finally, numerical simulations of proposed algorithms are presented and compared.KeywordsPacket LossKalman FilterFractional OrderVariable OrderPacket DropoutThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.