AbstractsBusiness Management & Administration

Bandwidth is one of the limited resources in Long Term Evolution (LTE) and LTE-Advanced (LTE-A) networks. Therefore, new resource allocation techniques such as the frequency reuse are needed to increase the capacity in LTE and LTE-A. However, the system performance is severely degraded using the same frequency in adjacent cells due to increase of intercell interference. Therefore, the intercell interference management is a critical point to improve the performance of the cellular mobile networks. This thesis aims to mitigate intercell interference in the downlink LTE and LTE-A networks. The first part of this thesis introduces a new intercell interference coordination scheme to mitigate downlink intercell interference in macrocell-macrocell scenario based on user priority and using fuzzy logic system (FLS). A FLS is an expert system which maps the inputs to outputs using “IF...THEN” rules and an aggregation method. Then, the final output is obtained through a deffuzifaction approach. Since this thesis aims to mitigate interference in downlink LTE networks, the inputs of FLS are selected from important metrics such as throughput, signal to interference plus noise ratio and so on. Simulation results demonstrate the efficacy of the proposed scheme to improve the system performance in terms of cell throughput, cell edge throughput and delay when compared with reuse factor one. Thereafter, heterogeneous networks (HetNets) are studied which are used to increase the coverage and capacity of system. The focus of the next part of this thesis is picocell because it is one of the important low power nodes in HetNets which can efficiently improve the overall system capacity and coverage. However, new challenges arise to intercell interference management in macrocell-picocell scenario. Three enhanced intercell interference coordination (eICIC) schemes are proposed in this thesis to mitigate the interference problem. In the first scheme, a dynamic cell range expansion (CRE) approach is combined with a dynamic almost blank subframe (ABS) using fuzzy logic system. In the second scheme, a fuzzy q-learning (FQL) approach is used to find the optimum ABS and CRE offset values for both full buffer traffic and video streaming traffic. In FQL, FLS is combined by q-learning approach to optimally select the best consequent part of each FLS rule. In the third proposed eICIC scheme, the best location of ABSs in each frame is determined using Genetic Algorithm such that the requirements of video streaming traffic can be met. Simulation results show that the system performance can be improved through the proposed schemes. Finally, the optimum CRE offset value and the required number of ABSs will be mathematically formulated based on the outage probability, ergodic rate and minimum required throughput of users using stochastic geometry tool. The results are an analytical formula that leads to a good initial estimate through a simple approach to analyse the impact of system parameters on CRE offset value and number of ABSs.