الفهرس 
Chapter 1: IntroductionOnly 14 pages are availabe for public view
 |  | from | 130 |  |  |
| | | from | 130 | | |
Abstract
The forecast of mobile challenge is that traffic produced by smartphones will eclipse more than it does today. Lately, smartphone traffic predicted to be rapidly grown by ten times and the total data traffic for all mobile devices by eight times. In addition, more than 90% of mobile traffic will come from smartphones.
For this reasons, mobile users will need to be actively move towards the 5^th generation of mobile (5G), whose architecture is based on the technology of 4^th generation (4G) mobile network. Furthermore, the 5G offers a higher connectivity speeds with less latency and time delays. Currently, the 5G of mobile technology is attractive and has the ability to change the field of wireless communication. With up to 1,000x data rates faster than 4G, The 5G technology changed the consumer behavior, as well as in smartphones, communications, IoT, gaming, and Augmented Reality / Virtual Reality (AR/VR).
Heterogeneous network (HetNet) indicates the use of multiple types of access nodes in a wireless network. A simple 5G HetNet is composed of one hexagonal gridded macro cellular mobile network with randomly deployed Pico base stations (PBSs) located within the macro cellular coverage area. In 5G, whatever the used technology, user equipment (UE) association technique is strongly required to decide to which network base station (BS) the UE is associated with, before starting data transmission connection. UE association acts as an indispensable part in improving data load balancing, efficient use of spectrum, and networks energy consumption efficiency. The challenge here is to make the appropriate association that achieves the minimum required data rate for each user with acceptable complexity and better operation for the network.
Game theory is one of the most powerful methods that have the ability to solve the network association problems and have different advantages in realization the fundamental interaction of many players. Multiple strategies combination to incorporate best strategies for all players is theoretically defined as equilibrium. Results of the applied game are to achieve Nash Equilibrium, so that no one of the players (UEs or BSs) has the ability to raise its own utility by modifying his strategy without decrease in utility of any other players. In this work, the players could be the BSs or the UEs or both, and strategies are set up from the related decisions of user association. The pragmatic UE association problem has been drafted as an optimized form of problem and resolved by the Nash Bargaining Solution (NBS). Results of the simulation of HetNet demonstrate that the algorithm enables network operators to encourage low allocation of network resources. This is because the model reduces the electrical power allocated to each cell and it is possible in some cases to shut down the cell completely. In addition, it ensures that users could be connected equitably by both macro and small cell. Also, it is predicted that this technique will provide an algorithm with low-complexity with a high performance guarantee.
One hypothesis of 5G cellular networks architecture is using three tiers; Macro base station (MBS), Wireless Fidelity (Wi-Fi) Pico BS, and Wireless Gigabit (WiGig) Femtocell BS (FBS). In this scenario, the needs to select the best BS to the UE to connect with become an emergency case. The challenge is to associate the UE to the best BS/tier to achieve the optimal operation for the network and maximum data rate for all UE. There are two proposed algorithms of cell association Cell Based Algorithm (CBA) and Tier Based Association Algorithm (TBA). The first algorithm CBA is related to finding the best BS in the three tiers and then associates the UE to one of them based on the association algorithm. The second proposed type TBA is related to deciding the best tier for the UE and then associating the UE to one of the BS of this tier based on the association algorithm. The opportunistic user association problem is formed as an optimization problem, which is solved by NBS. These two proposed algorithms achieve acceptable high System Data Rate (SDR) and fairness with low processing time with no effect on the network operation and mobility. In addition, this the proposed algorithms make an efficient comparison between suggested algorithms and other association algorithms in literature. The results of comparison show that the proposed algorithms achieve better system performance and results.
The SBS distribution effect on the overall performance of the cellular network is studied in the work. The homogeneous Poisson Point Process (PPP), Poisson Cluster Processes (PCP), and Matern Cluster Processes (MCP) are considered as distribution models used for this study. The cell association algorithm that is being used in this research is CBA, which is based on NBS. The challenges addressed feature the HetNets cell association work in three tiers with the perspective of newly emerging WiGig technology. Another challenge addressed in this work is the user distribution method consideration. Also, it is assumed that the user is distributed based on non-uniform user distribution model in which the user density depends on the distance to the associated BSs.