Properties of the characteristic function generator of the two-state Cauchy estimator

Yu Bai, Jason L. Speyer, Moshe Idan

Research output: Contribution to journalArticlepeer-review

Abstract

In order to better capture impulsive noises in dynamic systems, a state estimator in the presence of Cauchy distributed process and measurement noises has been studied in recent years. The Cauchy estimator is determined by expressing in closed form the characteristic function (CF) of the unnormalized conditional probability density functions of the states given measurement history. The CF is comprised of a sum of exponentials multiplied by a coefficient, both being nonlinear functions of the measurements and the spectral variable. In this paper, we uncover important properties of the exponential terms in the CF for the two-state Cauchy estimator. These properties can be used to simplify the estimator structure significantly.

Original languageEnglish
Pages (from-to)3639-3665
Number of pages27
JournalSIAM Journal on Control and Optimization
Volume57
Issue number6
DOIs
StatePublished - 2019

Keywords

  • Cauchy probability density function
  • Characteristic functions
  • Nonlinear estimation with heavy tailed noises

All Science Journal Classification (ASJC) codes

  • Control and Optimization
  • Applied Mathematics

Cite this