Gaussian Mixture Filters and Hybrid Positioning

Simo Ali-Loytty and Niilo Sirola

Abstract:	This paper presents, develops and compares Gaussian Mixture Filter (GMF) methods for hybrid positioning. The key idea of the developed method is to approximate the prior density as a Gaussian mixture with a small number of mixture components. We show why it is sometimes reasonable to approximate a Gaussian prior with a multicomponent Gaussian mixture. We also present both simulated and real data tests of different filters in different scenarios. Simulations show that GMF gives better accuracy than Extended Kalman Filter with lower computational requirements than Particle Filter, making it a reasonable algorithm for the hybrid positioning problem.
Published in:	Proceedings of the 20th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2007) September 25 - 28, 2007 Fort Worth Convention Center Fort Worth, TX
Pages:	562 - 570
Cite this article:	Ali-Loytty, Simo, Sirola, Niilo, "Gaussian Mixture Filters and Hybrid Positioning," Proceedings of the 20th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2007), Fort Worth, TX, September 2007, pp. 562-570.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In