An efficient framework for quality of service in wireless sensor network using machine learning

Abstract

Wireless Sensor Network is an emerging technology that has gained attention in a real-world. Due to the rapid growth of smart sensors, it is utilized in diverse applications of wireless networks such as intruder detection, transportation, IoT, smart cities, military, industrial, agricultural and health monitoring. It is difficult to assure the Quality of Service (QoS) in real-world applications due to changes in topology, limited resources and heterogenous traffic flow across the dynamic network. The optimized QoS can be achieved by improving its parameters such as maintainability, packet error ratio, reliability, scalability, availability, latency, jitter, throughput, priority, periodicity, deadline, security and packet loss ratio. It is a challenging job to achieve high performance in a real world because sensors are dispersed in an adverse environment. The performance parameters are classified into four parts: layered WSN architecture, deployment phase, measurability, application and network specific. Privacy and security levels include integrity, confidentiality, safety and security. newline