KFUPM ePrints

Dynamic Spectrum Allocation For Heterogeneous Cognitive Radio Networks With Femtocells

l Dynamic Spectrum Allocation For Heterogeneous Cognitive Radio Networks With Femtocells. Masters thesis, King Fahd University of Petroleum and Minerals.

[img]PDF - Accepted Version
Restricted to Abstract Only until 01 January 2015.
Available under License Creative Commons Attribution Non-commercial No Derivatives.

2249Kb

Arabic Abstract

الطيف الترددي هو المورد الأكثر ندرة في الاتصالات اللاسلكية. الراديو الذكي هو تقنية حديثة تم اقتراحها للتعامل مع مشكلة ندرة الطيف. أحد المشاكل الرئيسية التي قد تحد من انتشار شبكات الراديو الذكي هي ارتفاع أسعار الطيف الترددي. عدة بيئات تسويقية (مثل سوق الاحتكار) قد ينتج عنها ارتفاع في أسعار الطيف الترددي على مشغلي شبكات المستخدمين الفرعيين. في هذه الأطروحة، طريقة جديدة تقترح لتحسين فاعلية الطيف و تقليل تكاليف شرائه من شبكات الراديو الذكي. شبكة المستخدمين الفرعيين، ممثلة في محطة البث الذكية، تحدد عدد القنوات المطلوب شراؤها لخدمة مستخدميها قبل الشروع في شراء الطيف من المستخدمين الأساسيين. يمكن تحقيق هذا الأمر عن طريق تصنيف المستخدمين الفرعيين ضمن مجموعات وفقا للمسافات بينهم. تظهر النتائج التجريبية أن الطريقة المقترحة تقلل عدد القنوات المطلوب شراؤها من المستخدمين الأساسيين و بالتالي تعمل على تحسين فاعلية الطيف و تقليل كلفته على المستخدمين الفرعيين.

English Abstract

The frequency spectrum is the scarcest radio resource in wireless communications. Cognitive radio is an emergent technology which has been proposed to deal with the spectrum scarcity problem. One of the major problems that may limit the emergence of cognitive radio networks is high spectrum costs. Several marketing environments (e.g. Monopoly market) may result in high spectrum costs for the secondary user (SU) operators. In this work, a novel GPS-assisted scheme is proposed to enhance spectrum efficiency and reduce the operational cost of cognitive radio networks with femtocells. The SU network, represented by the cognitive base station (CBS), determines the minimum number of channels needed for its operation before it starts purchasing spectrum from the primary user (PU) networks. This can be achieved by grouping the femtocells into non-interfering groups based on the distances between them. %Grouping the femotcells will result in a reduction in the number of channels to be purchased/rented by the SU operator, and will consequently cut the operational costs of the SU network. Two approaches for implementing the grouping scheme are proposed; namely, the femtocell secondary user (FSU)-based and the femtocell base station (FBS)-based grouping algorithms. The complexity of each grouping algorithm is evaluated and both are compared in terms of the required time to simulate and the resulting number of spectrum channels to be purchased. Results showed that the FBS-based grouping is much less complex, but results in larger number of spectrum channels to be purchased. Additionally, the complexity of the update process for each method is derived, where results show that sorting the groups in an ascent order, according to the number of members in each group for FSU-based grouping and the category for FBS-based grouping, will generally reduce the complexity of the update process. Moreover, the outage performance of both the FSU-based and the the FBS-based grouping algorithms is compared, and it was shown that the FSU-based grouping algorithm results in a better outage performance (lower outage probability) than the FBS-based grouping one. In addition, several methods for optimizing and enhancing the grouping algorithms are proposed, and their performances have been evaluated. Results show that the CBS profit maximization algorithm achieves higher expected profits than the distance-based grouping with the minimization of the distance threshold, but with no QoS guarantees. Finally, an extension for the grouping scheme to the co-channel deployment scenario is presented, which helps to further reduce the number of purchased channels, but at the cost of worse outage performance.



Item Type:Thesis (Masters)
Subjects:Electrical
Divisions:College Of Engineering Sciences > Electrical Engineering Dept
Committee Advisor:Zummo, Salam
Committee Members:Mesbah, Wessam and Al-Ahmadi, Saad
ID Code:139284
Deposited By:YOUSEF NAIEF SHNAIWER (g201204420)
Deposited On:10 Jun 2014 08:28
Last Modified:10 Jun 2014 08:28

Repository Staff Only: item control page