Channel-aware downlink scheduling for quality of service in long term evolution network

• Main Author: Najim, Alaa Omer
• Format: Thesis
• Language: English
• Published: 2018
• Online Access: http://psasir.upm.edu.my/id/eprint/68727/1/FSKTM%202018%202%20IR.pdf; http://psasir.upm.edu.my/id/eprint/68727/

Long Term Evolution (LTE) is a recently evolving technology proposed by the Third Generation Partnership Project to provide a smooth migration toward the Fourth Generation of cellular networks. Radio Resource Management (RRM) is one of the core modules in LTE network architecture which includes functions to manage and handle involved system users who are communicating with their resources sites. In this thesis, Packet Scheduling is defined as one of the main RRM functions, and it is responsible for the intelligent selections of users and transmissions of their packets such that radio resources are efficiently utilized, and user’s Quality of Service (QoS) requirements are satisfied. The emphasis of this research is on the radio resource scheduling of LTE downlink. For flows with different volumes, delay has been defined as a major challenging issue that restricts the delivered network performance to subscribed users. In this thesis, a Delay-oriented Resource Allocation (DoRA) scheduler is proposed to provide an efficient allocation of wireless resources across users of different traffic volumes to minimize delay. Initially, a range of constraints was defined based on a recent algorithm from the literature, namely Delay Based Weighted Proportional Fair (DBWPF). These constraints comprise the use of an inefficient method to determine delay by the buffer size. In the proposed algorithm we, utilize Head of Line delay with channel-related parameters such as user data rate in addition to several pre-defined QoS tunable parameters to formulate a rule that prioritizes different user flows assigned with channel resources in order to improve delay. The proposed algorithm using system-level simulation experiments with performance comparison with DBWPF in terms of end-to-end delay, fairness, system throughput, Packet Loss Ratio (PLR), and spectral efficiency was evaluated. The output results indicate that the proposed DoRA scheduling algorithm outperforms DBWPF in providing a minimum level of end-to-end delay.