Effect of Signal Propagation Model Calibration on Localization Performance Limits for Wireless Sensor Networks

In this paper, we focus on wireless sensor network-based localization for user devices (UDs). Prior to UD localization, signal propagation model (SPM) is required to calibrate using training samples from a number of location grids. However, SPM usually suffers from measurement noise and inevitable e...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on wireless communications 2021-05, Vol.20 (5), p.3254-3268
Main Authors: Zhou, Bingpeng, So, Hing Cheung, Mumtaz, Shahid
Format: Article
Language:English
Subjects:
Calibration
Cramer-Rao lower bound
Design optimization
Error analysis
Internet-of-Things localization
Localization
Location awareness
Loss measurement
Lower bounds
Measurement uncertainty
Noise
Noise measurement
Path loss model calibration
Performance enhancement
performance limit
Propagation
Resource allocation
Training
Wireless communication
Wireless networks
Wireless sensor networks
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we focus on wireless sensor network-based localization for user devices (UDs). Prior to UD localization, signal propagation model (SPM) is required to calibrate using training samples from a number of location grids. However, SPM usually suffers from measurement noise and inevitable error in calibration grid (CG) locations. This will significantly degrade UD localization performance. Nevertheless, the impact of CG location error, CG layout and the number of CGs on SPM calibration performance has not been characterized. Furthermore, the effect of SPM calibration error on UD localization performance has not been developed. In this paper, we aim to provide a unified framework for performance analysis of SPM calibration and UD localization. Firstly, we establish a closed-form Cramér-Rao lower bound on SPM calibration error and UD localization error, respectively. Secondly, the impact of measurement noise, CG location error and the number of CGs on SPM calibration performance is revealed. Thirdly, the influence of SPM calibration error, CG location error and measurement noise on UD localization performance is studied. The effect of modeling mismatch is also studied. The obtained analysis framework builds a theoretical basis for the design of efficient system optimization strategies, including resource allocation and CG deployment optimization, for UD localization performance enhancement.
ISSN:1536-1276
1558-2248
DOI:10.1109/TWC.2020.3048744