Least square approximation technique for energy conservation in wireless sensor networks
Ishwer Kumar Bayer, Surendar Soni · 2014
In this paper, we present a method for regenerate and predict the values of sensors in a wireless sensor network without direct communication between sensor node and sink node based on least square approximation. The main disadvantage of Wireless Sensor Networks (WSNs) is their limited power source. The main power consumer in a sensor node is its radio transceiver. One of the most effective methods to reduce the data transmission rate is data prediction. Data prediction technique given in this paper can reduce amount of data transmitted, which results in energy saving and the overall network lifetime increases. Through simulation we have studied that how sink node not only predict the future values but also generate the initial values based on the parameter which will send to the sink node. We conducted our simulations on the actual data collected from 54 sensors deployed in the Intel Berkeley Research lab.