Per-survivor Processing-based Sequence Detection with Dependent Observations of Multiple Sensors in the Presence of Parameter Uncertainty

DÜLEK B., Isk S.

IEEE Signal Processing Letters, vol.30, pp.603-607, 2023 (SCI-Expanded) identifier

  • Publication Type: Article / Article
  • Volume: 30
  • Publication Date: 2023
  • Doi Number: 10.1109/lsp.2023.3277346
  • Journal Name: IEEE Signal Processing Letters
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, Applied Science & Technology Source, Communication Abstracts, Compendex, Computer & Applied Sciences, INSPEC, Metadex, Civil Engineering Abstracts
  • Page Numbers: pp.603-607
  • Keywords: Copula theory, hypothesis testing, parameter uncertainty, per-survivor processing, state sequence detection, temporal and spatial dependence
  • Hacettepe University Affiliated: Yes

Abstract

The problem of detecting the state sequence of a binary stochastic process with multiple sensors is considered. It is assumed that the sensors' observations are coupled both spatially and temporally. Copula theory is used to model the unknown spatial coupling across the observations of different sensors while the temporal dependency is taken into account using a first order Markov chain. A per-survivor processing-based algorithm is proposed to estimate the unknown parameter vector and determine the correct state sequence with reduced computational complexity. Numerical examples indicate that it is possible to keep track of the underlying state sequence while satisfactorily estimating the unknown parameters with the proposed method.