STATE-AWARE JOINT CHANNEL ASSIGNMENT AND ROUTING IN MULTI-RADIO MULTI-CHANNEL WIRELESS MESH NETWORKS BY OMAR ZAKARIA A thesis submitted in fulfilment of the requirement for the degree of Doctor of Philosophy in Engineering Kulliyyah of Engineering International Islamic University Malaysia OCTOBER 2015
24
Embed
STATE-AWARE JOINT CHANNEL ASSIGNMENT AND ROUTING IN …
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
STATE-AWARE JOINT CHANNEL ASSIGNMENT AND ROUTING IN MULTI-RADIO MULTI-CHANNEL
WIRELESS MESH NETWORKS
BY
OMAR ZAKARIA
A thesis submitted in fulfilment of the requirement for the degree of Doctor of Philosophy in Engineering
Kulliyyah of Engineering International Islamic University Malaysia
Wireless Mesh Network is one of the promising architecture for providing last-mile broadband Internet connectivity to network users. The network capacity in 802.11-based single channel wireless mesh network is highly affected by interference caused by backhaul wireless links’ transmissions. This makes it inadequate for the new deployment scenarios with high number of users and traffic demands. To increase the network capacity, mesh routers are equipped with multiple radio interfaces. As a consequence, various wireless links can simultaneously operate within a set of orthogonal channels instead of a single channel. Routing and channel assignment are fundamental challenges in such networks, where the two functions determine how the traffic distributes over different links and channels. Therefore, for a given traffic load distribution, both channel assignment and routing need to be efficiently determined. The interdependent nature of routing and channel assignment has attracted researcher’s attention to address these two issues jointly. In addition, re-configuration is required in dynamic traffic loads to ensure optimal network resources utilization. Frequent re-configuration degrades the network performance. This is because re-configuration of channels and routes disrupt the network traffic and increase the packet loss and delay. The main objective of this research is to develop an efficient joint state-aware algorithm, which is capable of adapting the traffic load variation with less traffic disruption. In developing the proposed solution, the re-configuration cost should be identified and considered. Firstly, the problem is formulated as a multi-objective optimization problem. The aim of this optimization problem is to minimize four objective functions, namely the maximum channel-link utilization, average network contention, channel re-assignment cost and re-routing cost. Then a heuristic algorithm called State-Aware Joint Routing and Channel Assignment (SA-JRCA) is proposed to address these challenges. The proposed algorithm is compared with the proposal of Avallone et al., 2013 and the proposal of Raniwala et al., 2004. The ns-2 simulator is used for evaluation. The proposed and compared works are evaluated and analyzed based on various metrics, such as maximum channel-link utilization, average network contention, channel re-assignment cost, re-routing cost, average throughput, and average end-to-end delay. The proposed algorithm shows better performance compared with the other two proposals. A new metric is proposed to evaluate the network performance. The proposed average network contention metric shows more correlations with network performance than maximum channel-link utilization. The results show that the proposed algorithm achieved the highest packet delivery ratio with more consistency with the traffic variation. In contrast, the other two algorithms show degradation in the performance with higher traffic variation and their achieved packet delivery ratio, reduced by 13%, 21% respectively when the traffic load varied from 10% to 50%.
iii
خلاصة البحثAbst
تعتبر الشبكات اللاسلكية المتشابكة واحدة من أهم الشبكات الواعدة لتأمين الاتصال بالإنترنت ذو النطاق الأخير. وتتأثر الطاقة الاستيعابية للشبكات اللاسلكية المتشابكة التقليدية العريض لمستخدمي الشبكة في الميل
ذي القناة الواحدة بشكل كبير بالتداخل الناشئ بين الوصلات الاسلكية لنواة الشبكة، وهذا يجعلها غير مناسبة دة سعة للسيناريوهات الجديدة، وخصوصاً مع تزايد عدد مستخدمي الشبكة واحتياجاتهم. ومن أجل ز�
استيعاب مثل هذه الشبكات تم تزويد كل من الموجهات المتشابكة بأكثر من راديو، و بذلك يمكن لعدة وصلات لاسلكية أن تعمل على أكثر من قناة في نفس الوقت عوضاً عن القناة الواحدة. وتعتبر عملية التوجيه وعملية
في هذه الشبكات، لأ�ا تؤثر على توزعّ حركة البيا�ت توزيع القنواة على الراديوهات واحدة من أكبر التحد�تعلى الروابط اللاسلكية وعلى القنوات المختلفة. وللحصول على الإعداد الأكفأ للشبكة لا بد من تحديد طرق التوزيع لكل حمولة حركية بالتوازي مع تحديد القنوات المخصصة للراديوهات، وبسبب هذا الترابط بين كل من
لتوجيه وعملية التوزيع للقنوات يعمل الباحثون على حل كلا المسألتين مع بعضهما البعض. بالإضافة إلى عملية اأنه ولضمان الاستثمار الأمثل لموارد الشبكة فإنه من الضروري إعادة توجيه وإعادة توزيع القنوات على
اء الشبكة بإعادة الضبط المتكررة للشبكة نتيجة الراديوهات بعد كل تغيرّ في الأحمال المرورية.مع العلم أنه يتأثر أدللتغير المستمر بالأحمال المرورية وذلك لأن عملية إعادة الضبط تؤدي إلى تعطيل مؤقت في حركة البيا�ت في الشبكة و تزيد من فقدان الحزم وتأخيرها. إن الهدف الرئيسي من هذا البحث هو تطوير خوارزمية فعالة واحدة
التوجيه وإعادة تحديد القنوات بحيث تكون قادرة على التكيف مع أي تغير في توزيع الحمولة لكل من إعادةوبأقل تعطيل ممكن في حركة البيا�ت. وللبدء بالحل تمّ تصميم المشكلة كمشكلة تحسينٍ متعددة الأهداف، وتم
ؤشرات (الأهداف) الأربعة تحديد تكلفة إعادة الضبط، وحلّ هذه المشكلة يتضمن العمل على إنقاص قيمة المالتالية: مؤشر أكبر استخدام للقنوات ومؤشر متوسط التنافسية في الشبكة ومؤشر تكلفة إعادة التعيين للقنوات
تعمل على حل المشكلة السابقة. ولقياس أداء وفعالية وجيه. ومن ثم تم اقتراح خوارزميةومؤشر تكلفة إعادة الت النموذج المقترح من قبل رانيوالا ومع ٢٠١٣ آخرون،عامأفالون و مقترحتها مع كل من الخوارزمية المقترحة تم مقارن
كبر�مج للمحاكاة، بر�مج محاكات الشبكات النسخة الثانية وفي سبيل ذلك تم استخدام .٢٠٠٤ وآخرون عامالأقصى للقنوات ومؤشر وتمّ التقييم اعتماداً على مقاييس كثيرة للأداء، ومن هذه المقاييس: مؤشر الاستخدام
متوسط التنافس في الشبكة ومؤشر تكلفة إعادة تعيين القنوات ومؤشر تكلفة إعادة التوجيه ومؤشر متوسط أداءً أفضل بالمقارنة مع النموذج المقترح الإنتاجية ومؤشر متوسط التأخير من النهاية إلى النهاية. وقد أظهر
كما بينت التجارب أن مؤشر متوسط التنافس في الشبكة أكثر ترابطا مع . النماذج الأخرى التي تم المقارنة بها أداء الشبكة من مؤشر الاستخدام الأقصى. كما أظهرت الطريقة المقترحة أن نسبة تسليم الحزم فيها هي الأعلى
ليم الحزم عند . في حين انخفضت نسبة تسارنة مع كل من المقترحين الآخرينمقوالأقل تأثراً عند تباين حركة المرور % ١٠ على التوالي عند تغيير نسبة التباين في الحمل المروري من% ٢١، %١٣ إلى المقترحين الآخرينكل من
.%٥٠ إلى
iv
APPROVAL PAGE
The thesis of Omar Zakaria has been approved by the following:
STATE-AWARE JOINT CHANNEL ASSIGNMENT AND ROUTING IN MULTI-RADIO MULTI-CHANNEL WIRELESS
MESH NETWORKS
I hereby affirm that the International Islamic University Malaysia (IIUM) holds all
rights in the copyright of this Work and henceforth any reproduction or use in any
form or by means whatsoever is prohibited without the written consent of IIUM. No
part of this unpublished research may be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means, electronic, mechanical, photocopying,
recording or otherwise without prior written permission of the copyright holder.
Affirmed by Omar Zakaria.
……………………………. ……………….. Signature Date
vii
I dedicated this thesis to my beloved family:
To my beloved Father
To my beloved Mother
To my beloved brother and sisters
To my lovely Wife
To my lovely sons
viii
ACKNOWLEDGEMENTS
All praise be to ALLAH s.w.t, the Most Beneficial and the Most Merciful. Best prayers and greetings to our holy prophet Muhammad p.b.u.h and his family.
First and foremost, Alhamdulillah, all praise be to Allah, for having mercy on me and giving me the health, patience, courage and determination to carry out this research.
I would like to express my sincere gratitude to my main supervisors, Prof. Dr. Aisha Hassan Abdalla Hashim for her supervision. I greatly appreciate all her dedication, guidance, advice and inspiration to improve the quality of this research and thesis. This thesis would not have been possible without her help and motivation.
I also would like to forward my gratitude to my co-supervisors, Prof. Dr. Othman O. Khalifa and Prof. Dr. Mohammad Azram. I greatly appreciate their guidance and advice throughout my research. With their continuous support and guidance, I am able to complete my PhD.
My appreciation is also extended to Assoc. Prof. Dr. Wan Haslina Hassan, Dr.Rashid Saeed and Dr. Khalid Hindawi. I am very thankful for their support and help extended to me during my PhD.
Many thanks also to the Malaysia Ministry of Higher education for the scholarship assistance in financing my study.
Special thanks to IIUM for providing a comfortable environment and sufficient resources and services to me.
I am also extremely grateful to my parents whose continuous blessing and sacrifices for everything in my life. I would especially like to thank my wife, for all her patience and her understanding to all the difficulties that I have encountered. Without the patience of my family, to whom I owe everything, it would have been impossible to accomplish my study.
Lastly, I like to thank my fellow friends for their assistance during my study. Thank you very much to Faiz, Lalitha, Mojtaba, Yunus, Waleed, Mahmoud, Mohammad, Ali, Yahia, Loay, Mistura, and Khalid.
ix
TABLE OF CONTENTS
Abstract .................................................................................................................... ii Abstract in Arabic .................................................................................................... iii Approval Page .......................................................................................................... iv Declaration ............................................................................................................... v Copyright Page ......................................................................................................... vi Table of contents ...................................................................................................... ix List of Tables ........................................................................................................... xii List of Figures .......................................................................................................... xiii List of Abbreviations ............................................................................................... xvi CHAPTER ONE: INTRODUCTION .................................................................. 1
1.1 Introduction............................................................................................... 1 1.2 Background of Research ........................................................................... 1 1.3 Problem Statement .................................................................................... 4 1.4 Research Hypothesis and Philosophy ....................................................... 5 1.5 Research Objectives.................................................................................. 6 1.6 Research Methodology ............................................................................. 6 1.7 Scope of Research..................................................................................... 8 1.8 Thesis Organization .................................................................................. 9
CHAPTER TWO: CHANNEL ASSIGNMENT AND ROUTING IN MULTI-RADIO MULTI-CHANNEL WIRELESS MESH NETWORKS ..................... 12
2.3.6.1 Network Model ................................................................... 29 2.3.6.2 Optimization Function and Fairness ................................... 30 2.3.6.3 Constraints .......................................................................... 32
2.3.7 Joint Channel Assignment and Routing Classification ................... 40 2.4 Related Works .......................................................................................... 44
2.4.1 Centralized Joint Channel Assignment and Routing ...................... 44 2.4.1.1 Static Channel Assignment and Routing ............................ 45 2.4.1.2 Static Channel Assignment with Link Scheduling and Routing ........................................................................................... 52 2.4.1.3 Dynamic Channel Assignment and Routing ...................... 54
3.1 Introduction............................................................................................ 74 3.2 System Model ........................................................................................ 74
3.2.1 Design Considerations and Assumptions ........................................ 75 3.2.2 Network Model ............................................................................... 76 3.2.3 Interference Model .......................................................................... 77
3.3 Mathematical Problem Formulation ...................................................... 78 3.3.1 Constraints ...................................................................................... 78
4.4 Simulation Results and Performance Analysis. ..................................... 115 4.4.1 The Impact of Route-variation Weighting Parameter 𝝎𝝎 ............... 122 4.4.2 The Impact of Number of Radios ................................................. 125 4.4.3 The Impact of Number of Channels .............................................. 129 4.4.4 The Impact of Traffic Variation .................................................... 134 4.4.5 The Impact of Network Traffic Load ............................................ 135