Supervised learning-based Lifetime Extension of Wireless Sensor Network Nodes

Jaewoong Kang, Jongmo Kim, and Mye Sohn+
 

Department of Industrial Engineering, Sungkyunkwan University, Suwan, Korea

{kjw1727, dignityc, myesohn}@skku.edu

 

 

Abstract

In this paper, we propose a new approach to increase the sustainability of the Wireless Sensor Network (WSN) nodes by extending their lifetimes. To do so, we attempt to find the optimal values of the collection interval and the transmission interval for each task that can maximize the lifetime of the WSN nodes by applying machine learning techniques. As a preprocessing for finding the optimal value of two parameters, we first determine the combination of nodes necessary to perform each task using the wrapper method. In addition, we applied Simulated Annealing (SA) to find the values of two parameter that lower power consumption without being significantly affected by the WSN’s performance. To prove the superiority of the method, we perform two kinds of experiments. Finally, we prove the reduction of energy consumption using our framework.

Keywords: Sustainability, Wireless Sensor Network, Interval Determination, Machine Learning

 

+: Corresponding author: Mye Sohn

InfoScience Laboratory, Department of Systems Management Engineering, Sungkyunkwan University,

2066 Seobu-ro, Jangan-gu, Suwon 440-746, Republic of Korea, Tel: +82-(0)31-290-7605

 

Journal of Internet Services and Information Security (JISIS), 9(4): 59-67, November 2019

DOI: 10.22667/JISIS.2019.11.30.059 [pdf]