|
Document Type |
: |
Thesis |
|
Document Title |
: |
QOS-BASED MOBILITY USING SDN/NFV IN WIRELESS NETWORK جودة الخدمة أثناء التنقل في الشبكات اللاسلكية القائمة على SDN/NFV |
|
Subject |
: |
Faculty of Computing and Information Technology |
|
Document Language |
: |
Arabic |
|
Abstract |
: |
By 2020, network data rates are expected to increase a thousand-fold due to the current explosion in mobile technology and widespread use of smart devices such as tablets and smartphones. The 5G wireless network is coming with promises to meet these high data rates requests and more. In the same vein, many solutions have been proposed as enabler keys for the 5G network such as infrastructure densification, Software Defined Network (SDN) and Network function virtualization (NFV).
Densification means constructing a large number of small cells in a specific area. These cells are including different wireless technologies (LTE, IEEE 802.11, etc). Small cells benefit the network because of the shorter distance between the access point and the mobile node will increase the data rate and minimize the signal attenuation and noise. This could guarantee a better connection, which goes along with the high data rate requests expectations by the year 2020. The downside of the densification is the frequent handoffs which represent a challenge for the network nodes. Software-defined networking is another promising solution for the 5G networks which control the network through centralization. Whilst, Network function virtualization is described as a complementary technology to the software-defined network which virtualizes the network functions in data centers.
This thesis studies the combination of infrastructure densification, software-defined network, and network function virtualization in order to reduce the handoff delays and failures in the upcoming 5G networks which will enhance the provided quality of service.
To achieve these goals, we went through three stages as follows. First, classifying the solutions of the mobility management of SDN/NFV-based wireless networks into three categories namely: NFV-based, SDN-based and SDN/NFV-based architectures. These categories have evaluated in terms of scalability, signaling cost, handling handoff failures issues, and maintain a balanced load in the system.
Second, proposing a hybrid clustering technique using K-mean and the genetic algorithm called PGA. PGA maintains high accuracy results in short execution time compared to K-mean and KGA algorithms. Third, applying the proposed PGA algorithm to create a clustered data-plane of an SDN-based wireless network. The clustering aims at reducing the time of the scanning phase of the handoff process by minimizing the number of access points that needed to be scan by a node. The results show lower delays and a reduced number of handoff failures compared to the un-clustered network. |
|
Supervisor |
: |
Dr. Nadine Akkari and |
|
Thesis Type |
: |
Doctorate Thesis |
|
Publishing Year |
: |
1440 AH
2019 AD |
|
Co-Supervisor |
: |
Dr. Ghadah Aldabbagh |
|
Added Date |
: |
Monday, May 20, 2019 |
|
Researchers
| ســوزان صالح باسلوم | Basloom, Suzan Saleh | Researcher | Doctorate | |
|