Particle MCMC for Bayesian Microwave Control

TL;DR

This paper presents a Bayesian inference approach using Particle MCMC for estimating radioelectric properties from electromagnetic scattering data, leveraging high-performance computing and advanced sequential Monte Carlo methods.

Contribution

It introduces a novel Particle Marginal Metropolis-Hastings method with Rao-Blackwellised SMC and Kalman filters for high-dimensional inverse electromagnetic problems.

Findings

Successful estimation of material properties on synthetic data

Demonstration of parallelization techniques for high-dimensional problems

Integration of data assimilation methods from geophysics into electromagnetic inverse problems

Abstract

We consider the problem of local radioelectric property estimation from global electromagnetic scattering measurements. This challenging ill-posed high dimensional inverse problem can be explored by intensive computations of a parallel Maxwell solver on a petaflopic supercomputer. Then, it is shown how Bayesian inference can be perfomed with a Particle Marginal Metropolis-Hastings (PMMH) approach, which includes a Rao-Blackwellised Sequential Monte Carlo algorithm with interacting Kalman filters. Material properties, including a multiple components "Debye relaxation"/"Lorenzian resonant" material model, are estimated; it is illustrated on synthetic data. Eventually, we propose different ways to deal with higher dimensional problems, from parallelization to the original introduction of efficient sequential data assimilation techniques, widely used in weather forecasting, oceanography, geophysics, etc.