LTE-A蜂窝网络中D2D通信资源分配机制的随机模型与性能评价研究

项目名称： LTE-A蜂窝网络中D2D通信资源分配机制的随机模型与性能评价研究

项目编号： No.61272168

项目类型： 面上项目

立项/批准年度： 2013

项目学科： 自动化技术、计算机技术

项目作者： 雷蕾

作者单位： 北京交通大学

项目金额： 83万元

中文摘要： 在LTE-A蜂窝网络中引入设备间直接通信（D2D）功能能够提高系统的频谱利用率并使得设备间的发现更加快速灵活。由于具有D2D功能的蜂窝网络拓扑既不同于传统蜂窝网络，也不同于传统多跳网络，因此需要建立新的性能评价模型对D2D通信所带来的性能增益进行评估并对资源分配机制的优化设计进行指导。目前关于LTE-A蜂窝网络中D2D链路资源分配及性能评价的研究刚刚起步，大都假设用户数目不变且具有无限积压业务。与上述研究不同，本项目基于业务随机动态到达的假设建立D2D通信的分组级和流级性能评价模型，通过排队论、随机Petri网、随机网络演算等理论工具，对支持D2D功能的LTE-A无线蜂窝网络的统计服务质量等进行性能分析，定量得出LTE-A无线蜂窝网络中引入D2D功能后在时延，吞吐量等方面取得的性能增益，并对各种考虑链路之间可能存在的相互干扰的资源分配机制进行性能分析和比较。

中文关键词： 终端直通；随机Petri网；；；

英文摘要： The introduction of Device-to-Device Communications in LTE-Advanced cellular networks can both improve resource utilization and enable flexible and fast neighbour device discovery. Since the topologies of D2D enabled cellular networks are different from those of traditional cellular networks and multi-hop networks, it is necessary to formulate new performance evaluation models to quantify the amount of performance gain achieved by D2D communications, and to provide guidelines for the optimizations of the resource allocation schemes. The researches on the resource management and performance evaluation of D2D enabled LTE-A cellular networks have just begun in recent years, and most existing works assume that the number of users in the system is fixed and each user has infinite backlog to transmit. Different from the above studies, this project formulates packet and flow level performance evaluation models based on stochastic and dynamic traffic arrival processes. The quality-of-service (QoS) performances of D2D enabled LTE-Advanced networks are analyzed with queueing theory, stochastic Petri Nets, and stochastic network calculas. The performance gains in terms of delay and throughput etc. achieved by the introduction of D2D function into LTE-Advanced networks are quantified, and the performances of various resourc

英文关键词： D2D；stochastic Petri net；；；