A Systematic Review of Quality of Service in Wireless Sensor Networks using Machine Learning: Recent Trend and Future Vision. (15th August 2021)

• Record Type: Journal Article
• Authors: Pundir, Meena; Sandhu, Jasminder Kaur
• Is Part Of: Journal of network and computer applications. Volume 188(2021)
• Publication Date: 2021-08-15
• Subjects: Wireless Sensor Network -- Quality of Service -- Machine Learning -- Layered architecture -- Network-specific parameters -- Application-specific parameters; Microcomputers -- Periodicals; Computer networks -- Periodicals; Application software -- Periodicals; Micro-ordinateurs -- Périodiques; Réseaux d'ordinateurs -- Périodiques; Logiciels d'application -- Périodiques; Application software; Computer networks; Microcomputers; Periodicals; 004.05; 004
• DOI: 10.1016/j.jnca.2021.103084
• Languages: English
• ISSNs: 1084-8045

Abstract

Abstract: Wireless Sensor Network (WSN) is used in different research areas such as military, industry, healthcare, agriculture, Internet of Things (IoT), transportation, and smart cities. The reason behind this increased usage is the rapid development of smart sensors. There is a challenging need to satisfy the Quality of Service (QoS) requirements in different applications due to the dynamic network condition, heterogeneous traffic flows and resource-constrained behaviour of sensor nodes. Optimizing the QoS in terms of performance, privacy and security levels is an open issue in the WSN. It has limited resources and is deployed in hostile environment where achieving high performance is difficult. This performance level is categorized into four subcategories: deployment phase, layered architecture, measurability, network and application specific parameter. Privacy and security levels are divided into four parameters: security, confidentiality, integrity and safety. A systematic review is presented in this paper based on QoS parameters in the light of Machine Learning (ML) techniques. It also provides a methodological framework for the performance parameters. This study presents a statistical analysis of the past ten years ranging from 2011 to 2021 on various ML techniques used for the QoS parameters. Finally, the author's vision is highlighted with some discussion on the open issues which forms the baseline for the future research directions. Graphical abstract: Image 1 Highlights: Uniquely categorized the QoS based on performance, privacy and security levels. Discussion of ML techniques used for all QoS parameters in the past decade. Statistical analysis of QoS parameters in the light of ML techniques is carried out. Energy efficiency was found to be the most prominent parameter for QoS enhancement. Vision of research for all the QoS parameters is also presented.