Document Details
|
Document Type |
: |
Thesis |
|
Document Title |
: |
Energy Efficient Load Balancing On-demand Multipath Routing Protocol-based Q-learning in MANET (QAOMDV) بروتوكول توجيه متعدد المسارات ذو كفاءة في استخدام الطاقة وموزع للحمل للشبكات اللاسلكية المخصصة للأجهزة المتنقلة باستخدام خوارزمية التعلم Q |
|
Subject |
: |
Faculty of Computing and Information Technology |
|
Document Language |
: |
Arabic |
|
Abstract |
: |
In MANET networks, nodes operate around their batteries, a heavy loaded node runs out of energy faster than others, causing the node to eventually die, partition the network, and lose data. Therefore, balancing the load between multiple routes will provide less energy consumption leading to an extended network lifetime. Most recent researches proposed enhancement for pre-existing multipath routing protocols such as AOMDV, adding different metrics for route selection such as routing time, buffer size, available bandwidth, residual energy, etc. However, researches gaps revolve around neglecting load distribution, exhaustion use of the same path, or neglecting nodes’ residual energy.
In this research, we’re proposing on-demand load balancing multipath routing protocol by combining the Online Q-learning algorithm with AOMDV multipath routing protocol to preserve nodes’ residual energy.
We conducted the simulation for our proposed routing protocol using Network Simulator-2 (ns2), the simulation’s results were evaluated against Quality of Service metrics: Packet Delivery Ratio, End to End delay, Average Energy Consumption, and Routing Overhead. The evaluation results obtained indicate that the proposed protocol outperformed in terms of Average Energy Consumption, Packet Delivery Ratio and Routing overhead. While it suffered from high end to end delay when compared to the state of art.
Key Words: Energy Efficient, MANET, Reinforcement Learning, Multipath Routing Protocol, Load Balancing |
|
Supervisor |
: |
Dr. Shahinda Sarhan |
|
Thesis Type |
: |
Master Thesis |
|
Publishing Year |
: |
1445 AH
2023 AD |
|
Added Date |
: |
Wednesday, October 4, 2023 |
|
Researchers
| مرام سلطان العتيبي | Alotaibi, Maram Sultan | Researcher | Master | |
|
Back To Researches Page
|