Citation: Li, S.; Welsen, S.; Brusic, V.
Multi-AP and Test Point Accuracy of
the Results in WiFi Indoor
Localization. Sensors 2022, 22, 3709.
https://doi.org/10.3390/s22103709
Academic Editor: Elena Simona
Lohan
Received: 8 March 2022
Accepted: 11 May 2022
Published: 12 May 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Multi-AP and Test Point Accuracy of the Results in WiFi
Indoor Localization
Shuyu Li
1
, Sherif Welsen
1,
* and Vladimir Brusic
2
1
Department of Electrical and Electronic Engineering, University of Nottingham Ningbo China (UNNC),
199 Taikang East Road, Ningbo 315100, China; shuyu.li@nottingham.edu.cn
2
School of Computer Science, University of Nottingham Ningbo China (UNNC), 199 Taikang East Road,
Ningbo 315100, China; vladimir.brusic@nottingham.edu.cn
* Correspondence: sherif.welsen@nottingham.edu.cn
Abstract:
WiFi-based indoor positioning has attracted intensive research activities. While localization
accuracy is steadily improving due to the application of advanced algorithms, the factors that affect
indoor localization accuracy have not been sufficiently understood. Most localization algorithms
used in changing indoor spaces are Angle-of-Arrival (AoA) based, and they deploy the conventional
MUSIC algorithm. The localization accuracy can be achieved by algorithm improvements or joint
localization that deploys multiple Access Points (APs). We performed an experiment that assessed
the Test Point (TP) accuracy and distribution of results in a complex environment. The testing space
was a 290 m
2
three-room environment with three APs with 38 TPs. The joint localization using
three APs was performed in the same test space. We developed and implemented a new algorithm
for improved accuracy of joint localization. We analyzed the statistical characteristics of the results
based on each TP and show that the local space-dependent factors are the key factors for localization
accuracy. The most important factors that cause errors are distance, obstacles, corner locations, the
location of APs, and the angular orientation of the antenna array. Compared with the well-known
SpotFi algorithm, we achieved a mean accuracy (across all TPs) improvement of 46%. The unbiased
joint localization median accuracy improved by 20% as compared to the best individual localization.
Keywords: channel state information; indoor localization accuracy; joint localization; WiFi
1. Introduction
The rapid development of wireless communication technology and the omnipresence
of mobile devices enable the rapid growth of Location-Based Services (LBSs) [
1
,
2
]. Outdoor
LBSs, such as the Global Navigation Satellite System (GNSS), are used for applications
such as navigation, emergency services, timekeeping, or uses for military or geodesy
purposes
[3,4].
Indoor LBSs complement outdoor LBSs. For example, indoor LBSs can be
used to track assets, build management, provide indoor location information for emergency
services, and navigate customers in shopping centers [
5
–
7
]. In healthcare, applications
include monitoring patients in nursing homes, tracking Alzheimer’s patients, monitoring
the activities and movements of rehabilitating patients, or improving the safety of elderly
patients [8–10].
The outdoor LBSs cannot be used indoors due to blocked signals. This is because of
the additional challenges such as multipath effects, Non-Line-of-Sight (NLoS), moving
humans or objects, ambient noise, and electromagnetic interference, which need to be
addressed [11]. Reliable Indoor Positioning Systems (IPSs) must satisfy key requirements:
Accuracy: Accuracy is a key improvement target for most indoor LBSs’ research [
1
].
Improving the indoor LBSs can be addressed by signal processing, additional hardware,
model-based analysis, or new communication protocols [12–17].
Availability: IPSs should cover all locations within the serviced indoor area at all
times. Preferably, IPSs should work with widely available devices, such as mobile phones
Sensors 2022, 22, 3709. https://doi.org/10.3390/s22103709 https://www.mdpi.com/journal/sensors
