1. UKM Learning and Research Repository
2. Master Thesis / Tesis Master
3. Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Title:	Comparison approach system-level fault diagnosis in mobile ad hoc network using non-overlapping clustering technique
Authors:	Jarrah Hazim (P50110)
Supervisor:	Rosilah Hassan, Prof. Dr.
Keywords:	Wireless communication systems.
Issue Date:	28-Mar-2011
Description:	Mobile ad hoc network (MANET) is increasingly becoming the basis for civilian as well as military communication with a wide range of applications, and therefore, there is an urgent need to diagnose, manage, and deal with faults in such networks. System-level fault diagnosis, a one step of fault tolerant, is focused on detecting and identifying faulty nodes in MANET. Many approaches were proposed to achieve this goal. The Comparison approach, in particular, is believed to be one of the most effective and promising. In this approach, nodes in the network are assigned the same task. Each node performs the task and reports the result. By comparing the results from the different nodes, faulty nodes are detected. Although there are many models that have been proposed based on this approach, Chessa and Santi model seems to be the most relevant as it is the only model for ad-hoc networks which considered the communication shared nature of MANET. In this research, we provide a review of the known protocols and algorithms in the literature, which are based on Chessa and Santi model. We also provide a comprehensive comparison of these protocols and illustrate their limitations. We propose a new protocol based on a non-overlapping clustering technique. The proposed protocol can diagnosis faults in MANET even if its topology is dynamic. The correctness and complexity proof for our protocol is provided. We implement the protocol using OMNeT++ simulator, and the results show that our protocol has smaller system overhead.,Master / Sarjana
Pages:	77
Call Number:	TK5103.2.J365 2011
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

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