ENSEMBLED ADABOOST MACHINE LEARNING ALGORITHM WITH NONLINEAR REGRESSION TREE FOR ENERGY AWARE DATA GATHERING IN WSN

Abstract
Data gathering is a process of collecting more number of data from distributed sensor nodes and sends these data to sink node. During the data gathering, energy consumption (EC) is a major concern for enhancing the network lifetime (NL). Several WSN architectures have been developed to resolve this problem In order to improve the data gathering efficiency, AdaBoost Nonlinear Regression Tree Classification (ABNRTC) technique is developed. ABNRTC technique improves the data gathering with minimal EC. Initially, energy of each senor nodes is measured. After that, mobile sink node gathers the sensed information from the high energy sensor nodes with minimum delay. Then the mobile sink node classifies the collected data packet using nonlinear regression tree based on their relationship among the sensor nodes in WSN. The relationship between the data packets are measured using population Pearson product-moment correlation coefficient. AdaBoost algorithm is a boosting technique for grouping the several weak nonlinear regression tree classifiers to make a single final output of boosted classifier. Finally, the classified data packets are sent to base stations (BS). Simulation of ABNRTC technique is carried out with different parameters such as EC, NL, data gathering efficiency, delay, classification accuracy (CA), false positive rate (FPR) and classification time (CT).

Authors
S S Aravinth, A Arul Prasath, B Arunkumar
Dhirajlal Gandhi College of Technology, India

Keywords
Wireless sensor network, Nonlinear Regression Tree, Pearson Product-Moment Population Correlation Coefficient, AdaBoost Algorithm
Published By :
ICTACT
Published In :
ICTACT Journal on Communication Technology
( Volume: 10 , Issue: 3 )
Date of Publication :
October 2019
DOI :

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.