Wireless Sensor Network with Regression Analysis Based Improved A-Star Algorithm

Abstract

Wireless Sensor Network (WSN) technology has taken a big leap in recent times. We are advancing towards the use of WSN in our day to day life. WSN is composed of portable sensor nodes. The sensor nodes are the pillars of network. Sensor Nodes gather data from the environment and transmit this data to the sink node; from where this data is transmitted to Control Station (CS) for further processing over the internet. Though this small sensor node comes with various advantages but there are limitations as the sensor node has limited energy, memory and computation proficiency. Here, the proposed work is an improved A-star algorithm to increase the network lifetime. A Regression Analysis based function has been used to calculate the Heuristic value for selecting a participating node in the routing process. The selection is based upon the following variables – Packet Reception Ratio (PRR), Residual Energy and Node Buffer State. The sink node selects a node based on this heuristic value calculated using Regression Analysis out of these parameters for current routing process. By using this method, all nodes sustain the full network life as a node with lesser energy level is not opted for the participation in routing process until unless all other nodes do not reach to the energy level of this node with lesser energy (or threshold value).