Ensembled adaboost learning with id3 algorithm for energy aware data gathering in WSN

Abstract

In this paper, data collection is the operation of gathering a lot of details from the sensor nodes and shipping it to the sink node. The use of Network is increasingly required to perform these processes, so, increases energy consumption. A lot of WSN architectures are designed to solve this complex problem. By using a technique called Decision Tree Classifier using Adaboost (DTCA) algorithm, can extension that data collection efficiency, as well as reducing the delay and Power consumption. In the proposed methodology, the power of each sensor node should be estimated at the outset. Then the mobile sink node receives the information from the high power sensor nodes with minimal delay. The mobile sink node classifies the data pockets using the Decision Tree Classifier. This classifies based on the relationship between the sensor nodes in WSN. That relationship is measure using the method of population Pearson product moment correlation coefficient. Adaboost algorithm is a combination of several weak non-linear classifiers to create a higher classification. Then finally, it sends classified particulars to Base Station. The operation of the DTCA system is convey out with divergent parameters such as classification time, EC, (Network Lifetime) NL, data collection capability, Classification Accuracy (CA), (FPR) false positive rate and delay.