SMIER: An SVM and MCDA Based, Intelligent Approach for Enhanced Reliability in Wireless Sensor Networks

Volume 6 Issue 3 May - July 2017

Research Paper

SMIER: An SVM and MCDA Based, Intelligent Approach for Enhanced Reliability in Wireless Sensor Networks

Mohammad Samadi Gharajeh*, Maryam Askari Zivayeki**, Sareh Askari***

* Young Researchers and Elite Club, Tabriz Branch, Islamic Azad University, Tabriz, Iran.

** Department of Computer Engineering, Bostanabad Branch, Islamic Azad University, Bostanabad, Iran.

*** Department of Computer Engineering, Tabriz Branch, Islamic Azad University, Tabriz, Iran.

Gharajeh, M. S., Zivayeki, M. A., Askari, S. (2017). SMIER: An SVM and MCDA Based, Intelligent Approach For Enhanced Reliability In Wireless Sensor Networks. i-manager’s Journal on Communication Engineering and Systems, 6(3), 1-8.https://doi.org/10.26634/jcs.6.3.13611

Abstract

Reliability is one of the big challenges in Wireless Sensor Networks (WSNs). It can be improved by using some of the intelligent and knowledge-based mechanisms, such as Support Vector Machine (SVM) and Multiple Criteria Decision Analysis (MCDA). This paper proposes an SVM and MCDA based, intelligent approach for enhanced reliability in WSNs, called SMIER. It is considered for a cluster-based sensor network in a way that every Cluster-Head (CH) selects one of its Non Cluster-Head (NCH) nodes as a backup node in a period of time. Initially, the suggested SVM algorithm determines failure probability of each NCH node based on number of events and average distance to events. Afterward, the suggested MCDA controller calculates success rate of the NCH node by using three parameters, including remaining energy, distance, and failure probability. Simulation results show that the proposed approach surpasses some of the existing works in terms of packet delivery ratio, number of alive nodes, and average remaining energy.