Spontaneous network establishment routing algorithm to enhance energy efficiency in wireless sensor network using machine learning

Abstract

Wireless Sensor Network (WSN) is the interconnection between Things / objects embedded with Hardware and software. In WSN Small end devices (like sensors) and high end devices (like servers) are connected to the internet. RPL routing protocol was designed in WSN for efficient and reliable data collection in low power and lossy networks. The RPL (Routing Protocol for Low-Power and Lossy Networks) network uses objective functions such as hop count or ETX (Expected Transmission Count) for construction and maintenance of network topology. The DODAG (Destination Oriented Directed Acyclic Graph) construction is designed using Rank property give rise to problems such as greedy parent election, thundering head and network instability. WSN enabled in Software Defined Network (SDN), the routers are controlled using a controller server node. It is a dynamic network due to the presence of mobile nodes and energy constrained nodes. The routing is the process of detecting route from source to target. In dynamic networks like WSN routing is a challengeable task. Therefore, the following methods are presented to overcome the above problems in WSN. Reliable Parent Selection Using Advanced Reinforcement Learning Method. A Dynamic network to Enhance Quality of Service in SDN Based WSN (DTEQT) Using Machine Learning In the first method, an energy-efficient reliable parent selection is provided in RPL. The optimum parent is selected based on the predicted energy of the parent using RSS. The predication is carried out using advanced Reinforcement Learning method. newline