Quality of Service and Channel Aware Packet Scheduling Algorithm for Long-Term Evolution System

摘要

As the development of wireless networks has evolved, Long-Term Evolution (LTE) was brought in to fulfill the demands and needs of users. This is especially true in the area of the wireless network's Quality of Service (QoS) in real-time applications. There are various solutions for problems in the downlink packet scheduling, but there is still room to have a higher QoS to provide greater satisfaction to users. A better user experience requires real-time applications to have a minimum amount of delay. A study of recent existing scheduling strategies implemented in LTE and an overview table is provided for overall comparison and analysis is presented. Based on the literature review, further research is performed on Exponential Rule (EXP Rule) scheme. Thus, this research have proposed several schemes whereby were experimented and analysed. The proposed schemes include 1) EXP Rule with MT (EXP_MT Rule), 2) Simplified EXP Rule (sEXP Rule), 3) Modified Exponential Rule (mEXP Rule), and 4) Enhanced EXP Rule with MT (E2M Rule). Here, E2M Rule performs better than the benchmarked EXP Rule and outstands the other proposed schemes, thus is further experimented in a two-level scheduling framework. In this research, two-level scheduling framework that pays attention to realtime applications,which include video and Voice over Internet Protocol (VoIP), is proposed and developed.