1. UKM Learning and Research Repository
2. PHD Thesis / Tesis PHD
3. Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina

Title:	Quality-of-service based scheduling schemes for smart grid and mobile users over Long Term Evolution-Advanced networks
Authors:	Ayman Hussein Hajjawi (P75370)
Supervisor:	Mahamod Ismail, Prof. Dr.
Keywords:	Networks Wireless Scheduling schemes Bandwidth Universiti Kebangsaan Malaysia -- Dissertations
Issue Date:	26-Oct-2017
Description:	The latest wireless mobile standard, Long Term Evolution-Advanced (LTE-A) Release 13, was developed by the 3rd Generation Partnership Project to support users with very high data rate and low delay. LTE-A network is a potential wireless backbone for the smart grid because of its reliability and stability. However, the network performance significantly depends on the bandwidth resource allocation among smart grid users. Moreover, LTE-A scheduler has not been designated for smart grid communications, leading to unfair bandwidth allocation among users specifically to users with real-time applications. Although the current scheduling algorithms show acceptable performance for specific applications, their performance sharply drops when implemented to scenarios with diverse applications, resulting in a high Packet Drop Rate (PDR) and low spectral efficiency. In addition, the complexity of these algorithms causes a high delay in the scheduling decision, exceeding the LTE-A Transmission Time Interval (TTI) of 1 ms. In addition, the LTE-A scheduling decision is not adapted to the dynamic application's preferences. The main objective of this research is to propose new Quality-of-Service (QoS)-based scheduling schemes for mobile and smart grid applications over LTE-A networks. The schemes are classified as Generic Priority (GP) and Particular Priority (PP). The GP scheduler allocates the bandwidth resources fairly among mobile and smart grid applications, whereas the PP scheduler is proposed to meet the critical delay requirements of the smart grid applications. The GP scheduler is comprised of three algorithms: Bandwidth Resources Estimation for Classes (BREC), Bandwidth Resource Allocation for Users (BRAU), and Multi-Decision (MD). BREC algorithm is proposed to improve the bandwidth allocation fairness and throughput among mobile and smart grid applications using proportional allocation technique. BRAU algorithm is proposed to improve the bandwidth allocation fairness among the users in each application. MD scheduling algorithm is adapted to prioritize users according to their dynamic preferences to reduces PDR and maintains the complexity low. The PP scheduler consists of two algorithms: PDR-Based Allocation (PBA) and Preferences-Based Scheduling (PBS). PBA algorithm allocates the bandwidth resources to the applications according to PDR based on self-learning technique. PBS algorithm schedules the users according to the users' preferences. The proposed algorithms are validated using LTE-Sim under normal and overloaded traffic conditions. The mobile and smart grid applications are uniformly distributed across the cell. The performance is evaluated and compared with three conventional schedulers: Modified-Largest Weighted Delay First, Exponential Rule (EXP-Rule), and Proportional Fairness in terms of PDR, fairness, spectral efficiency, delay, and throughput. The BREC algorithm improved throughput up to 14% and 30% for mobile and smart grid users, respectively, compared with EXP-Rule scheduler. Additionally, BRAU improved the fairness up to 20% for mobile users and 15% for smart grid users. MD reduced PDR to approximately 15% and 25% for mobile and smart grid users, respectively. Moreover, PP scheduler reduced the PDR of smart grid users to 36%. However, the proposed algorithms need further improvement to support diverse applications with different QoS requirements such as Internet of Things environment.,Ph.D.
Pages:	131
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.