Switching Extended Kalman Filter Bank for Indoor Localization Using Wireless Sensor Networks

Jung Min Pak

doi:10.3390/electronics10060718

Abstract

This paper presents a new filtering algorithm, switching extended Kalman filter bank (SEKFB), for indoor localization using wireless sensor networks. SEKFB overcomes the problem of uncertain process-noise covariance that arises when using the constant-velocity motion model for indoor localization. In the SEKFB algorithm, several extended Kalman filters (EKFs) run in parallel using a set of covariance hypotheses, and the most probable output obtained from the EKFs is selected using Mahalanobis distance evaluation. Simulations demonstrated that the SEKFB can provide accurate and reliable localization without the careful selection of process-noise covariance.

Highlights

Localization involves tracking locations of objects that are of interest, such as robots, humans, vehicles, and equipment [1,2,3]
Time of arrival (TOA) [5], time difference of arrival (TDOA) [6], and angle of arrival [7] are typical wireless measurements used for indoor localization
Because wireless measurements of wireless sensor networks (WSNs) are typically represented by nonlinear measurement models, nonlinear stochastic filters such as the extended Kalman filter (EKF) and the particle filter (PF) are often used for indoor localization [8,9,10,11]

Summary

Introduction

Localization involves tracking locations of objects that are of interest, such as robots, humans, vehicles, and equipment [1,2,3]. Because wireless measurements of WSNs are typically represented by nonlinear measurement models, nonlinear stochastic filters such as the extended Kalman filter (EKF) and the particle filter (PF) are often used for indoor localization [8,9,10,11]. Uncertainty in the motion model is one of the oldest problems in the field of target tracking To solve this problem, interacting multiple model (IMM) filtering [16] was developed. This paper proposes a new filtering algorithm, the switching extended Kalman filter bank (SEKFB), to overcome the uncertain process-noise covariance problem. Without the need for the careful selection of Q, the SEKFB algorithm performs accurate localization compared to the best achieved performance of an EKF.

Indoor Localization Scheme and Proposed Algorithm

Simulation

Finish

Findings

Conclusions