Designing an Optimal Sensor Network via Minimizing Information Loss

TL;DR

This paper presents a novel sensor placement method that minimizes information loss by integrating physics-based simulations with Bayesian experimental design, validated through a temperature monitoring case study.

Contribution

The paper introduces a new model-based sensor placement criterion and an efficient optimization algorithm that leverages physics simulations and Bayesian principles for optimal sensor network design.

Findings

Outperforms random sampling in sensor placement accuracy.

Effective with limited sensors in complex environments.

Validated on air temperature monitoring in Phoenix.

Abstract

Optimal experimental design is a classic topic in statistics, with many well-studied problems, applications, and solutions. The design problem we study is the placement of sensors to monitor spatiotemporal processes, explicitly accounting for the temporal dimension in our modeling and optimization. We observe that recent advancements in computational sciences often yield large datasets based on physics-based simulations, which are rarely leveraged in experimental design. We introduce a novel model-based sensor placement criterion, along with a highly-efficient optimization algorithm, which integrates physics-based simulations and Bayesian experimental design principles to identify sensor networks that "minimize information loss" from simulated data. Our technique relies on sparse variational inference and (separable) Gauss-Markov priors, and thus may adapt many techniques from Bayesian experimental design. We validate our method through a case study monitoring air temperature in Phoenix, Arizona, using state-of-the-art physics-based simulations. Our results show our framework to be superior to random or quasi-random sampling, particularly with a limited number of sensors. We conclude by discussing practical considerations and implications of our framework, including more complex modeling tools and real-world deployments.