Integrated combined layer algorithm of jamming detection and classification in manet / Ahmad Yusri Dak

The latest technologies in Mobile Ad-Hoc Network (MANET) allows networking infrastructure to be set up quickly and easily. Every node in MANET plays a role as a router that creates remote network access regardless of time or location. As such, the dynamic nature of the infrastructure opens MANET to...

Full description

Saved in:

Bibliographic Details
Main Author:	Dak, Ahmad Yusri
Format:	Thesis
Language:	English
Published:	2019
Subjects:	Indexes (General)
Online Access:	https://ir.uitm.edu.my/id/eprint/57290/2/57290.PDF
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-uitm-ir.57290
record_format	uketd_dc
spelling	my-uitm-ir.572902024-02-14T01:10:15Z Integrated combined layer algorithm of jamming detection and classification in manet / Ahmad Yusri Dak 2019-04 Dak, Ahmad Yusri Indexes (General) The latest technologies in Mobile Ad-Hoc Network (MANET) allows networking infrastructure to be set up quickly and easily. Every node in MANET plays a role as a router that creates remote network access regardless of time or location. As such, the dynamic nature of the infrastructure opens MANET to being vulnerable to jamming attacks. A study found that eighty one percent of attacks occurs at the physical and MAC layers of the protocol stack. This happened due to inadequate guidelines using standard parameters to detect jamming attacks. In addition, the use of techniques such as watchdogs, routing and databases cannot solve problems which is detecting an important layer thoroughly, lack of parameters in the database and failure to detect some important jammers due to limited jammer detection methods. Therefore, knowing the types and characteristics of jammers will enable researchers to develop defence techniques that can prevent jammer attacks. However, due to limited study conducted in the field of jammer classification, the defence phase has become impossible. Presently, the classification accuracy of the assessment is proportional to the size of the data. Increasing the size of the data will make it more accurate. However, in many cases it is not possible to get such large amounts of data due to certain constraints. Therefore, there is a need to find a new valuation method that does not require large data sets yet yields the same results. Thus, this study proposes the development of algorithms to detect and classify jamming attacks using a set of parameters on the physical layer and MAC in MANET. In addition, the methodology to determine the accuracy of the dataset based on limited amount of data will be evaluated. The methodology consists of five main stages. The first stage is to apply reverse engineering method to obtain the specific patterns of individual jammers. This creates the jamming detection and classification parameters. The second stage is detecting jammers by integrating both lower layers by developing Integrated Combined Layer Algorithm (ICLA). The third stage is to classify individual jammers according to the specific pattern and characteristics design as defined in jamming identification and classification parameters. It involves development of Max-Min Rule-Based Classification Algorithm. The fourth stage is to design evaluation methodology of Max-Min Rule-Based Classification Algorithm using classifier model. Finally, both algorithms are validated against the findings in various literatures. Two types of testing were conducted to evaluate the effectiveness of proposed algorithms, which is the performance and accuracy test. Performance tests were conducted against jammers' model at MAC and physical layer. The findings show a significant increase in the detection rate up to 96.26% and 69.15% of deceptive jammer and reactive jammer using ICLA compared to literature. In addition, methodologies for jamming classification to measure the accuracy for each classifier were developed and tested using SVM and WEKA. Results show that data with normalization and scaling for polynomial kernel with cross validation presented the highest accuracy assessment which is 76.142% compared to other classifiers. Therefore, jammer attack detection and classification, studies were provided with better knowledge and technology to develop more robust defence techniques. 2019-04 Thesis https://ir.uitm.edu.my/id/eprint/57290/ https://ir.uitm.edu.my/id/eprint/57290/2/57290.PDF text en public phd doctoral Universiti Teknologi MARA Faculty of Computer and Mathematical Sciences Mohd Khalid, Noor Elaiza (Dr.)
institution	Universiti Teknologi MARA
collection	UiTM Institutional Repository
language	English
advisor	Mohd Khalid, Noor Elaiza (Dr.)
topic	Indexes (General)
spellingShingle	Indexes (General) Dak, Ahmad Yusri Integrated combined layer algorithm of jamming detection and classification in manet / Ahmad Yusri Dak
description	The latest technologies in Mobile Ad-Hoc Network (MANET) allows networking infrastructure to be set up quickly and easily. Every node in MANET plays a role as a router that creates remote network access regardless of time or location. As such, the dynamic nature of the infrastructure opens MANET to being vulnerable to jamming attacks. A study found that eighty one percent of attacks occurs at the physical and MAC layers of the protocol stack. This happened due to inadequate guidelines using standard parameters to detect jamming attacks. In addition, the use of techniques such as watchdogs, routing and databases cannot solve problems which is detecting an important layer thoroughly, lack of parameters in the database and failure to detect some important jammers due to limited jammer detection methods. Therefore, knowing the types and characteristics of jammers will enable researchers to develop defence techniques that can prevent jammer attacks. However, due to limited study conducted in the field of jammer classification, the defence phase has become impossible. Presently, the classification accuracy of the assessment is proportional to the size of the data. Increasing the size of the data will make it more accurate. However, in many cases it is not possible to get such large amounts of data due to certain constraints. Therefore, there is a need to find a new valuation method that does not require large data sets yet yields the same results. Thus, this study proposes the development of algorithms to detect and classify jamming attacks using a set of parameters on the physical layer and MAC in MANET. In addition, the methodology to determine the accuracy of the dataset based on limited amount of data will be evaluated. The methodology consists of five main stages. The first stage is to apply reverse engineering method to obtain the specific patterns of individual jammers. This creates the jamming detection and classification parameters. The second stage is detecting jammers by integrating both lower layers by developing Integrated Combined Layer Algorithm (ICLA). The third stage is to classify individual jammers according to the specific pattern and characteristics design as defined in jamming identification and classification parameters. It involves development of Max-Min Rule-Based Classification Algorithm. The fourth stage is to design evaluation methodology of Max-Min Rule-Based Classification Algorithm using classifier model. Finally, both algorithms are validated against the findings in various literatures. Two types of testing were conducted to evaluate the effectiveness of proposed algorithms, which is the performance and accuracy test. Performance tests were conducted against jammers' model at MAC and physical layer. The findings show a significant increase in the detection rate up to 96.26% and 69.15% of deceptive jammer and reactive jammer using ICLA compared to literature. In addition, methodologies for jamming classification to measure the accuracy for each classifier were developed and tested using SVM and WEKA. Results show that data with normalization and scaling for polynomial kernel with cross validation presented the highest accuracy assessment which is 76.142% compared to other classifiers. Therefore, jammer attack detection and classification, studies were provided with better knowledge and technology to develop more robust defence techniques.
format	Thesis
qualification_name	Doctor of Philosophy (PhD.)
qualification_level	Doctorate
author	Dak, Ahmad Yusri
author_facet	Dak, Ahmad Yusri
author_sort	Dak, Ahmad Yusri
title	Integrated combined layer algorithm of jamming detection and classification in manet / Ahmad Yusri Dak
title_short	Integrated combined layer algorithm of jamming detection and classification in manet / Ahmad Yusri Dak
title_full	Integrated combined layer algorithm of jamming detection and classification in manet / Ahmad Yusri Dak
title_fullStr	Integrated combined layer algorithm of jamming detection and classification in manet / Ahmad Yusri Dak
title_full_unstemmed	Integrated combined layer algorithm of jamming detection and classification in manet / Ahmad Yusri Dak
title_sort	integrated combined layer algorithm of jamming detection and classification in manet / ahmad yusri dak
granting_institution	Universiti Teknologi MARA
granting_department	Faculty of Computer and Mathematical Sciences
publishDate	2019
url	https://ir.uitm.edu.my/id/eprint/57290/2/57290.PDF
_version_	1794191832187404288

Integrated combined layer algorithm of jamming detection and classification in manet / Ahmad Yusri Dak

Similar Items