Simultaneous Localization and Mapping Problem in Wireless Sensor Networks

TL;DR

This paper presents an online method for simultaneous localization and mapping in wireless sensor networks that accounts for environmental dynamics using a probabilistic perturbation model and Sequential Monte Carlo methods.

Contribution

It introduces a novel online EM algorithm for real-time SLAM in wireless sensor networks, incorporating environmental perturbations into propagation maps.

Findings

Effective localization accuracy demonstrated in Monte Carlo experiments

Robustness to environmental changes shown through adaptive modeling

Real-time performance achieved with online inference algorithms

Abstract

Mobile device localization in wireless sensor networks is a challenging task. It has already been addressed when the WiFI propagation maps of the access points are modeled deterministically. However, this procedure does not take into account the environmental dynamics and also assumes an offline human training calibration. In this paper, the maps are made of an average indoor propagation model combined with a perturbation field which represents the influence of the environment. This perturbation field is embedded with a prior distribution. The device localization is dealt with using Sequential Monte Carlo methods and relies on the estimation of the propagation maps. This inference task is performed online, i.e. using the observations sequentially, with a recently proposed online Expectation Maximization based algorithm. The performance of the algorithm are illustrated through Monte Carlo experiments.