multiple links in different scenarios. Instead, to evaluate the performance of a network containing different scenarios, multiple drops should be run, one for each scenario, and merged afterward. This approach can be justified for conventional point-to-point cellular systems, where an end-to-end transmission involves only one link. However, this approach cannot be used in D2D systems, where an end-to-end transmission involves multiple heterogeneous links simultaneously. By introducing the UE's moving track and the location of clusters/scatters, COST2100 channel model seems to be one possible solution for this challenge. However, COST2100 is not designed for D2D channels and thus significant modifications are necessary.
紧接着上述挑战的另一个重要的挑战是，如何实现一个可以同时运行多个D2D场景的信道模型。如表1所示,D2D模型应该依靠不同的场景来描述不同的链接。到目前为止,3GPP D2D信道模型是唯一由3 GPP集团正式宣布遵循WINNER模型的基本建模方法。因此当前实现的3gpp D2D信道模型不允许同时在不同的场景中模拟多个链接。相反,为了评估包含不同场景的网络性能,应该运行多个单独以及合并后的场景。这种方法可以判断传统的从一个端到另一端的传输只有一个链接的点对点蜂窝系统是否合理的。然而,这种方法不能用于从一个端到端的传输同时涉及多个异构链接的D2D系统。通过对问题的跟踪以及引入散射/集群的移动位置,COST2100信道模型似乎是这一挑战一种可能的解决办法。然而,COST2100不是专为D2D信道设计的,所以大量的修改是必要的。
Actually, an ideal channel model would be closer to ray-tracing-based model, for example, deterministic ray-tracing channel model or GBSM. Considering the complexity issue and the ability to cover various communication scenarios, GBSM expresses obvious advantages over deterministic ray-tracing model. As mentioned in Section 3, COST2100 actually belongs to MBPGM. Therefore the development of D2D GBSMs may be better solution of the aforementioned two challenges. Taking into account the complexity aspect and the nature of propagation channels, a new D2D channel modelling approach is desirable. The new D2D channel modelling approach should combine the advantages of both MBPGM and GBSM. The development of such new D2D modelling approach is possible but challenging. COST 2100 model can be considered as an attempt which properly incorporates the UE's moving track and the location of clusters/scatters into conventional MBPGM. To better design this new modelling approach, it is important to find a proper way of introducing the scatter/cluster location model and the transition model among different propagation scenarios into current MBPGM.
实际上,一个理想的信道模型接近基于射线追踪的模型,例如,确定性射线跟踪模型或GBSM。考虑涵盖各种通信场景的复杂性问题时,GBSM表达明显优于确定性射线跟踪模型。正如在第3部分提及的,COST2100模型实际上属于MBPGM模型。因此发展D2D GBSMs可能是解决前面提到的两个挑战较好的方案。考虑到传播渠道的复杂性以及本质,可以采取一个新的D2D信道建模方法。新的D2D信道建模方法应结合MBPGM和GBSM的优点。这种新的D2D造型方法的发展是可能的但具有挑战性。COST2100模型尝试把跟踪到的问题以及集群/散射移动的位置与传统MBPGM的正确合并。更好地设计这种新的建模方法,重要的是在不同传播场景下找到一种引入散射/集群位置的模型并且把这种模型过渡到当前MBPGM。
To solve the aforementioned challenges, we have to face some other key challenges regarding channel measurement and channel data analysis and post-processing. For example, how to properly build up measurement campaign to measure various D2D scenarios while keeping the spatial consistency, how to measure multiple D2D pairs communication scenarios and
the impact to each other, how to obtain the accurate location of clusters/ scatterers from measured data etc. All these challenges necessitate innovation and breakthrough in the fields of channel measurements and channel data analysis and post-processing.
为了解决上述挑战,我们不得不面对一些关于信道测量、信道数据分析和后期处理的关键挑战。例如,如何在保持空间一致性的同时正确测量各种D2D场景,如何衡量多个D2D通信场景以及它们的相互影响,如何从测量数据中获得集群/散射的精确位置等。所有这些挑战需要在信道测量、信道数据分析和后期处理领域有创新和突破。 Conclusions 结论
In this paper, we put together a comprehensive overview of state-of-the-art D2D channel research, in order to facilitate channel-oriented D2D communication system design and optimisation. We first summarised D2D channel measurements in more than ten application scenarios, and provided in-depth discussion on their associated key parameters. Then, we critically reviewed and compared a number of prevalent channel models and their feasibility in D2D scenarios. We also used GBSM as an example to demonstrate certain channel characteristics for various D2D environments. Future challenges in this research area are also discussed in detail.
在本文中,我们建立一个最先进的综合概述D2D信道研究,为了方便channel-oriented D2D通信系统设计和优化。我们首先总结D2D信道测量在十多个应用程序场景,并提供深入讨论相关的关键参数。然后,我们回顾并比较严重的普遍在D2D场景中信道模型及其可行性。我们也使用GBSM作为一个例子来演示某些信道特性对各种D2D环境。在这个研究领域未来的挑战也详细讨论。 Acknowledgments 鸣谢
This work was jointly supported by the National 973 project (Grant no. 2013CB336700), the National Natural Science Foundation of China (Grant no. 61101079, 61222105, and 61471268), the Science Foundation for the Youth Scholar of Ministry of Education of China (Grant no. 20110001120129), the National 863 Project (Grant No.2014AA01A706), the State Key Laboratory of Rail Traffic Control and Safety (RCS2014K008, CS2014ZT11 and RCS2011ZZ002), Beijing Jiaotong University, and the project [13510711000] of the Science and Technology Commission of Shanghai Municipality “System design and demo-construction for cooperative networks of high-efficiency 4G wireless communications in urban hot-spot environments”.
这项工作是共同支持的国家973项目(批准号2013 cb336700),中国国家自然科学基金(批准号61101079,61101079,61471268),青年学者的科学基金会中国教育部(批准号20110001120129)、国家863项目(格兰特No.2014AA01A706),轨道交通控制与安全国家重点实验室(RCS2014K008 CS2014ZT11和RCS2011ZZ002),北京交通大学,项目[13510711000]上海市科委的“系统设计和demo-construction合作高效4 g无线通信网络在城市热点环境”。