Research On Packet Scheduling Algorithm Of Hsdpa In Wcdma System

High Speed Downlink Packet Access (HSDPA) is the evolution of WCDMA, proposed in 3GPP Release 5. The main object of HSDPA is to decrease the propagation delay in physical layer, increase the peak data transfer rate of user equipment (UE), and improve the spectral efficiency of downlink and the throughput of the whole cell. Based on the 5MHz bandwidth of previous WCDMA system, HSDPA provides increased downlink packet data throughput by means of many key techniques, such as adaptive modulation and coding, hybrid automatic repeat request and fast packet scheduling.HSDPA supports various kinds of services such as streaming, interactive and background. Heterogeneous services have different quality of service (QoS) requirements, for example, the requirements of maximum packet delay, the maximum acceptable packet drop rate, the minimum data rate. There are three issues for HSDPA Mac-hs scheduler design. One is to guarantee QoS requirements for heterogeneous services, another is to guarantee the fairness between users, the third is to maximize the utilization of the wireless resource. All of these depend on the packet scheduling algorithm.Firstly, the thesis gives an overview of the characteristics of HSDPA and the requirement on scheduling, and analyses the physical layer and MAC layer of HSDPA. Then we summarize the principles and the concepts of wireless packet scheduling algorithms, analyze and compare the traditional algorithms, including the real-time and non-real-time scheduling algorithms. Based on that, this thesis proposes a modified packet scheduling algorithm for multi-service scenarios on High Speed Downlink Packet Access (HSDPA).In this packet scheduling algorithm, we consider both the delay constraint and the fairness for each traffic class while guaranteeing different quality of service (QoS) for each traffic class. The strategy is evaluated in a scenario including steaming and interactive and background traffic. Numerical results show that the system performance can be significantly improved employing our algorithm than that with PF or M-LWDF method. Finally, some theoretical analysis and comparison for these algorithms are presented with some computer simulation results.