Backhauling in Heterogeneous Cellular Networks: Modeling and Tradeoffs
Backhauling in Heterogeneous Cellular Networks: Modeling and Tradeoffs
复制标题
DOI：
10.1109/twc.2015.2403321
复制
发表时间：
2015-06-01
影响因子：
10.4
通讯作者：
Kountouris, Marios
中科院分区：
文献类型：
--
作者：
Chen, Daniel C.;Quek, Tony Q. S.;Kountouris, Marios
过去几年，消费者对数据的需求大幅增长，小型蜂窝网络越来越被认为是科普这一需求的关键技术之一。与传统蜂窝网络相比，传统宏蜂窝网络内的小小区网络部署正在产生大量的异构性。然而，回程链路通常是系统性能和成本方面的瓶颈。在本文中，我们考虑了异构蜂窝网络中的回程问题，并提出了一个分层网络模型，使用叠加的独立齐次泊松点过程。我们得出的总预期延迟，考虑到无线链路上的重传，以及从有线和无线回程所产生的回程延迟。对于总的预期部署成本，我们考虑了基础设施成本和有线回程部署的建设成本。具体来说，我们能够使用我们简单易行的模型来描述延迟和部署成本的行为。此外，我们提出了一个基于延迟的访问控制策略，它可以提供更好的延迟性能，特别是在密集部署。我们的理论框架提供了一个基本的理解有线和无线回程之间的权衡和部署成本和异构蜂窝网络中的系统性能的影响。
Consumer demand for data has increased tremendously over the last years, and small cell networks are increasingly being considered as one of the key technologies to cope with this demand. Small cell network deployments within the conventional macro cellular networks are creating a significant amount of heterogeneity compared to traditional cellular networks. Nevertheless, the backhaul link is often the bottleneck in terms of system performance and cost. In this paper, we consider the backhaul issue in heterogeneous cellular networks and propose a hierarchical network model using superimposed independent homogeneous Poisson point processes. We derive the total expected delay by taking into account retransmissions over the wireless link, as well as the backhaul delay incurred from both wired and wireless backhaul. For the total expected deployment cost, we take into account infrastructure cost and the construction cost of wired backhaul deployment. Specifically, we are able to characterize the behavior of delay and deployment cost using our simple and tractable model. Furthermore, we propose a delay-based access control policy, which can provide better latency performance especially in dense deployments. Our theoretical framework provides a fundamental understanding of the tradeoff between wired and wireless backhaul and the effect on deployment cost and system performance in heterogeneous cellular networks.