الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 111 |  |  |
| | | from | 111 | | |
Abstract
Recently, the exponential growth of the numerous wireless devices and the data hunger applications have earned huge significance. This required imperious expansion of the 5G network to support the forthcoming 5G use cases such as video live-streaming, conferencing and online gaming. Moreover, the 5G cellular networks is planned to hugely elevate the capacity and the spectrum efficiency with respect to 4G. This can be achieved by utilizing Heterogeneous Networks (HetNets) which can enhance the system data rates and the Quality of Service (QoS) of the users as the Small Cells (SCs) are deployed within the Macro Cells (MCs) coverage area. Furthermore, SCs offer the benefit of providing service to previously uncovered region, and in the network regions demanding larger capacity.
It is foreseen that the massive growth in the SC deployment will be continued in the coming years leading to various challenges such as the interference among the SCs, their elevated power consumption and the elevated operating expenses. Thus, it is crucial to face those challenges to ameliorate the performance of 5G HetNets
One of the promising solutions to tackle the power consumption problem and ameliorate the Power Efficiency (PE), is to switch the SCs to sleep mode. Moreover, to solve the interference problem, Soft Frequency Reuse (SFR) algorithm is proposed. In SFR, every SC is split into center and edge regions where one of unutilized sub-bands by the edge regions of the adjoining SCs is allocated to the edge region of the SC while the remaining sub-bands are utilized in the center region of the SC having reduced transmission power to diminish the interference to the adjoining SCs.
The main objectives of this thesis are to: first present a new SFR-based scheme utilizing the Interference Contribution Rate (ICR) concept for the SC switching to mitigate the interference and to improve the PE in irregular 5G HetNets. After allocating the resources to the User Equipments (UEs), the SC switching is initiated based on the ICR concept. The on/off switching decisions of the SCs are determined according to the ICR values of the SCs to mitigate the interference between the SCs and to minimize the traffic losses. Second to propose a new approach that merge the usage of both Binary Particle Swarm Optimization (BPSO) algorithm with linear increasing Inertia Weight (IW) and SFR to maximize the PE of the SCs and minimize the number of active SCs while guaranteeing the QoS for the UEs. The proposed algorithm utilizes the linear increasing IW-BPSO algorithm for selecting the SCs to be switched on/off and propose the SFR utilizing Classification Trees (CTs) to enhance the network PE, taking into consideration the irregular nature of the 5G HetNets. The on/off switching of SCs using BPSO algorithm is carried out first; then the SFR is applied for sub-band allocation, to minimize the number of operations required for allocating the sub-bands to the SCs.
The results demonstrate that the proposed algorithms surpass the other conventional algorithms with regard to the total system throughput, the PE and the outage probability.