Quality-of-Service Prediction Techniques for Wireless Sensor Networks
Meena Pundir, Jasminder Kaur Sandhu, Ajay Kumar
Abstract
Abstract The rapid development of smart sensor nodes generates great interest in Wireless Sensor Network (WSN). The distinct applications of these networks have evolved in the field of border surveillance, Internet of Things, agriculture, healthcare, and disaster monitoring. It is a complex task to satisfy the need for Quality-of-Service (QoS) in a real-world environment, due to the size and dynamic nature of the mobile sensor nodes. This paper addresses the QoS metric based on the layered architecture of WSN. There are two types of layered architectures, namely classic-layered architecture and cross-layered architecture. In a classic-layered architecture, QoS is achieved based on the individual layer. Due to the interconnected nature of these layers, it is difficult to achieve the overall QoS in a classic-layered architecture. This challenge is resolved with the help of cross-layered architecture. Further, the proposed statistical analysis illustrated the fact that the QoS metric enhances the performance of the network in terms of latency, throughput, energy, and reliability. Finally, the machine learning approaches are discussed in the light of QoS metrics to enhance the overall performance of the network.