Please use this identifier to cite or link to this item:
Title: Clustering and Change Point Detection in Wireless Sensor Networks for Fire Detection Application
Researcher: Urvashi, Chugh
Guide(s): Singh, Yashwant
Keywords: Change Point Detection
Wireless Sensor Network
University: Jaypee University of Information Technology, Solan
Completed Date: 2016
Abstract: Wireless Sensor Networks WSNs are gaining magnitude day by day. This thesis focuses on to solve two major objectives of WSNs i to reduce power consumption in WSN operation and ii To make WSN usable for fire detection application. First objective aimed at to present a solution for energy saving in WSN operation via Clustering. Clustering is grouping of sensors which improve network life time by saving battery life of sensor nodes. This thesis work has given first contribution of a new clustering protocol named Mutual Exclusive Distributive Clustering MEDC. MEDC protocol selects cluster heads on remaining energy factor. It will select cluster head to that sensor node which one will be having highest residual energy under that range of communication. Next contributory work is to merging of MEDC with Hybrid Energy Efficient Distributive Clustering HEED protocol which results new clustering protocol named as Mutual Exclusive Hybrid Energy Efficient Distributive Clustering MEHEED which results in saving of energy. The second objective of thesis is change point detection. MEDC will select Cluster Heads after which selected cluster heads will perform first level Change Point Detection. Cluster heads will use fuzzy logic approach for change point detection. Fire detection application is considered here as example to show change point detection. It is assumed that sensors will sense three parameter heat index relative humidity and carbon monoxide. Cluster heads which are chosen by MEDC will collect and aggregate these sensed parameters from their cluster member nodes. Based on the aggregated values fire detection will be performed using fuzzy logic approach on cluster heads. Our algorithms have been simulated on MATLAB.Simulation of MEDC and MEHEED has been measured in terms of network lifetime with assumption that sensors are homogeneous type. MEHEED have improved network life time over MEDC over HEED. Next part of thesis is Change point detection simulation which is performed using FIS tool of MATLAB. newline
Appears in Departments:Department of Computer Science Engineering

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File192.09 kBAdobe PDFView/Open
02_declaration.pdf381.27 kBAdobe PDFView/Open
03_certificate.pdf174.49 kBAdobe PDFView/Open
04_acknowledgement.pdf119.42 kBAdobe PDFView/Open
05_table of contents.pdf75.78 kBAdobe PDFView/Open
06_list of tables and figures.pdf129.71 kBAdobe PDFView/Open
07_chapter 1.pdf190.1 kBAdobe PDFView/Open
08_chapter 2.pdf232.09 kBAdobe PDFView/Open
09_chapter 3.pdf469.02 kBAdobe PDFView/Open
10_chapter 4.pdf417.15 kBAdobe PDFView/Open
11_chapter 5.pdf492.38 kBAdobe PDFView/Open
12_chapter 6.pdf59.52 kBAdobe PDFView/Open
13_conclusion.pdf102.7 kBAdobe PDFView/Open
14_publications.pdf103.84 kBAdobe PDFView/Open

Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.