基于HSDPA的中继调度算法研究

发布时间：2018-10-24 08:40

【摘要】：随着UMTS系统的演进,HSDPA技术已经成为增强UMTS系统下行链路性能的主要手段。中继技术由于能够提高系统性能包括系统吞吐率和系统覆盖等已经被列入LTE-A标准化进程。鉴于此,HSDPA标准也在开始讨论是否通过引入中继来提高系统性能。但是中继的引入给传统的HSDPA系统调度带来新的问题,因此本论文重点讨论HSDPA中继系统中的调度问题,包括实时业务和非实时业务。在传统的HSDPA系统中,对于实时业务的调度,目前主流的调度算法是修正的最大加权时限优先(M-LWDF)算法,虽然可以继续采用该算法,但由于中继的引入,中继用户的延时增加必然会导致中继用户的丢包率增大,而且由于采用的是Type I中继,中继可以独立调度该中继下的用户,因此其时延并不是一个确定值,而是一个随机变量。此外,由于基站到中继的链路性能较好,可以将多个用户的分组聚合到同一个传输块中进行传输,此时基站需要确定分组聚合的策略。通过对以上问题的分析,本文提出了一种改进的M-LWDF算法,来设计中继的调度优先级。该调度优先级计算中不仅考虑到中继用户第二跳的时延和分组聚合还考虑到中继用户在基站端急需传输的分组数。仿真中,通过VoIP业务对该算法进行了验证,结果表明该算法能够在保证用户QoS要求的前提下,提高系统吞吐量。对于非实时业务的主流调度算法是比例公平调度算法,该算法可以在吞吐率和公平性之间取得很好的折中。但当HSDPA系统引入中继后,采用FDM中继无法调整中继回程链路在无线帧中所占的时间比例,因此该算法无法保证基站用户和中继用户之间的公平性。鉴于此,本文通过分析系统公平性与系统无线资源分配的关系,提出了一种增加中继优先级修正因子以保证全局公平性的调度方案,并通过系统级仿真对方案的性能进行了分析。仿真结果表明,该算法在系统吞吐率和公平性两个方面都有较好的性能表现。
[Abstract]:With the evolution of UMTS system, HSDPA technology has become the main means to enhance the downlink performance of UMTS system. Relay technology has been included in the LTE-A standardization process for its ability to improve system performance, including system throughput and system coverage. In view of this, the HSDPA standard is also beginning to discuss whether to introduce relay to improve system performance. However, the introduction of relay brings new problems to the traditional HSDPA system scheduling. Therefore, this paper focuses on the scheduling problems in HSDPA relay systems, including real-time services and non-real-time services. In the traditional HSDPA system, for real-time traffic scheduling, the current mainstream scheduling algorithm is the modified maximum weighted time-first (M-LWDF) algorithm, although it can continue to be used, but due to the introduction of relay, The increase of relay delay will inevitably lead to the increase of packet loss rate of relay users. Because the relay uses Type I relay, the relay can independently schedule the users under the relay, so the delay is not a definite value, but a random variable. In addition, because the link performance of base station to relay is good, the packets of multiple users can be aggregated into the same transmission block for transmission. In this case, the base station needs to determine the policy of packet aggregation. By analyzing the above problems, an improved M-LWDF algorithm is proposed to design the priority of relay scheduling. In this scheduling priority calculation, not only the delay of the second hop of the relay user and the aggregation of packets are taken into account, but also the number of packets that the relay user needs to transmit at the base station is considered. In the simulation, the algorithm is verified by VoIP service, and the result shows that the algorithm can improve the system throughput under the premise of guaranteeing the user's QoS requirement. The main scheduling algorithm for non-real-time traffic is proportional fair scheduling, which can achieve a good compromise between throughput and fairness. But when the relay is introduced into the HSDPA system, the FDM relay can not adjust the time ratio of the relay backhaul link in the wireless frame, so the algorithm can not guarantee the fairness between the base station user and the relay user. In view of this, by analyzing the relationship between system fairness and system wireless resource allocation, this paper proposes a scheduling scheme to increase relay priority correction factor to ensure global fairness. The performance of the scheme is analyzed by system-level simulation. Simulation results show that the proposed algorithm performs well in both throughput and fairness.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5

【相似文献】