Conjugate directions based order recursive implementation of post-Doppler adaptive target detectors

Abstract

An implementation for the post-Doppler adaptive target detectors enabling an efficient change of the subspace dimension is described. The proposed implementation uses the order recursive structure of the conjugate directions method and does not present any additional computational burden on the processor. The implementation can be particularly useful for the adaptive detectors with an indeterminate number of auxiliary vectors for the clutter covariance matrix estimation. Through the proposed method, the subspace dimension can be easily increased or decreased according to the multiplicity of the auxiliary vectors at no loss of computational efficiency.

Gaussian process‐based Bayesian non‐linear filtering for online target tracking

“…Estimating the internal state of a dynamic system from noisy observations is a common problem in the field of statistical signal processing [1–3], which finds various applications in navigation and guidance system such as radar tracking [4], sonar ranging [5] and satellite localisation [6]. One of the most popular techniques for state estimation are the Bayesian filtering methods in the context of stochastic dynamics systems, which computes the posterior probability density of the target system state recursively based on the Bayes’ theorem [7].…”