Bayesian likelihood-free localisation of a biochemical source using multiple dispersion models

TL;DR

This paper introduces a likelihood-free Bayesian approach that uses multiple dispersion models to accurately localize biochemical sources in the atmosphere, addressing the challenges of unknown likelihood functions and model variability.

Contribution

It presents a novel likelihood-free Bayesian method that integrates multiple dispersion models for biochemical source localization, enhancing robustness and applicability.

Findings

Successfully localized sources using experimental datasets

Demonstrated effectiveness of the likelihood-free approach

Compared performance across different dispersion models

Abstract

Localisation of a source of a toxic release of biochemical aerosols in the atmosphere is a problem of great importance for public safety. Two main practical difficulties are encountered in this problem: the lack of knowledge of the likelihood function of measurements collected by biochemical sensors, and the plethora of candidate dispersion models, developed under various assumptions (e.g. meteorological conditions, terrain). Aiming to overcome these two difficulties, the paper proposes a likelihood-free approximate Bayesian computation method, which simultaneously uses a set of candidate dispersion models, to localise the source. This estimation framework is implemented via the Monte Carlo method and tested using two experimental datasets.