Outlier detection in indoor localization and Internet of Things (IoT) using machine learning
Article
Article Title | Outlier detection in indoor localization and Internet of Things (IoT) using machine learning |
---|---|
ERA Journal ID | 5080 |
Article Category | Article |
Authors | Bhatti, Mansoor Ahmed (Author), Riaz, Rabia (Author), Rizvi, Sanam Shahla (Author), Shokat, Sana (Author), Riaz, Farina (Author) and Kwon, Se Jin (Author) |
Editors | Bhatti, Mansoor Ahmed |
Journal Title | Journal of Communications and Networks |
Journal Citation | 22 (3), pp. 236-243 |
Number of Pages | 8 |
Year | 2020 |
Place of Publication | Korea |
ISSN | 1229-2370 |
1976-5541 | |
Digital Object Identifier (DOI) | https://doi.org/10.1109/JCN.2020.000018 |
Web Address (URL) | https://ieeexplore.ieee.org/abstract/document/9143576 |
Abstract | In Internet of things (IoT) millions of devices are intel- ligently connected for providing smart services. Especially in in- door localization environment, that is one of the most concerning topic of smart cities, internet of things and wireless sensor net- works. Many technologies are being used for localization purpose in indoor environment and Wi-Fi using received signal strengths (RSSs) is one of them. Wi-Fi RSSs are sensitive to reflection, re- fraction, interference and channel noise that cause irregularity in signal strengths. The irregular and anomalous RSS values, used in a Wi-Fi indoor localization environment, cannot define the location of any unknown node correctly. Therefore, this research has de- veloped an outlier detection technique named as iF_Ensemble for Wi-Fi indoor localization environment by analyzing RSSs us- ing the combination of supervised, unsupervised and ensemble ma- chine learning methods. In this research isolation forest (iForest) is used as an unsupervised learning method. Supervised learning method includes support vector machine (SVM), K-nearest neigh- bor (KNN) and random forest (RF) classifiers with stacking that is an ensemble learning method. For the evaluation purpose accu- racy, precision, recall, F-score and ROC-AUC curve are used. The evaluation of used machine learning method provides high accu- racy of 97.8 percent with proposed outlier detection methods and almost 2 percent improvement in the accuracy of localization pro- cess in indoor environment after eliminating outliers. |
Keywords | Internet of things, localization, outliers, outliers de- tection |
ANZSRC Field of Research 2020 | 460609. Networking and communications |
Byline Affiliations | University of Azad Jammu and Kashmir, Pakistan |
Raptor Interactive, South Africa | |
Kangwon National University, Korea | |
Institution of Origin | University of Southern Queensland |
https://research.usq.edu.au/item/q6v9v/outlier-detection-in-indoor-localization-and-internet-of-things-iot-using-machine-learning
Download files
Published Version
Outlier_detection_in_indoor_localization_and_Internet_of_Things_IoT_using_machine_learning.pdf | ||
License: CC BY-NC 4.0 | ||
File access level: Anyone |
83
total views71
total downloads0
views this month0
downloads this month