Improved D/TA and Information Fusion Based on HMM Indoor Localization
Authors:
- 1 August 2013
- journal article
- Published by Trans Tech Publications, Ltd. in Applied Mechanics and Materials
- Vol. 373-375, 916-921
Abstract
This paper describes an efficient Bayesian framework for localization based on Ultra-wide Bandwidth (UWB) system. Approximate grid-based method based on the Hidden Markov Model (HMM) is an effective method to estimate the position of the Moving Terminal (MT) with the mixed line-of-sight/non-line-of-sight (LOS/NLOS) situation. This article proposes an algorithm by modifying the Position Transition Probability (PTP) according to the practical dynamic model and uses the information fusion effectively. We compare the Maximum Likelihood (ML) estimation with Detection/Tracking Algorithm (D/TA) estimation and its improved algorithm by simulation, in which the localization to an identical trajectory has been tested. The results of the analysis show that the proposed method has better accuracy and stability.Keywords
This publication has 10 references indexed in Scilit:
- Robust Relative Location Estimation in Wireless Sensor Networks with Inexact Position ProblemsIEEE Transactions on Mobile Computing, 2012
- Differential radio map-based robust indoor localizationEURASIP Journal on Wireless Communications and Networking, 2011
- CRLBs for WSNs localization in NLOS environmentEURASIP Journal on Wireless Communications and Networking, 2011
- A Jump Markov Particle Filter for Localization of Moving Terminals in Multipath Indoor ScenariosIEEE Transactions on Signal Processing, 2008
- Hidden Markov Models for Radio Localization in Mixed LOS/NLOS ConditionsIEEE Transactions on Signal Processing, 2007
- Mobile positioning using wireless networks: possibilities and fundamental limitations based on available wireless network measurementsIEEE Signal Processing Magazine, 2005
- Kalman tracking for mobile location in NLOS situations2004
- Mobile location tracking with NLOS error mitigation2004
- A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian trackingIEEE Transactions on Signal Processing, 2002
- Hidden Markov model for multidimensional wavefront trackingIEEE Transactions on Geoscience and Remote Sensing, 2002