面向5G的毫米波时变信道建模与仿真技术研究

第三、四代（3G、4G）移动通信信道模型是基于几何基础上的随机模型，由于路损经验模型与信道参数的统计特性，在特定的链路设计与仿真中时常会与实际信道有大的偏差。 毫米波传输是第五代（5G）通信无线接入与回传的关键技术，由于毫米波典型场景存在大量的移动散射体，所以时变信道特性与建模是5G信道研究的方向和热点。本项目基于国际电联（ITU）5G毫米波候选频段，在5G典型场景开展实际信道测试，获得时变信道数据，把人工神经网络和随机微分方程等研究方法引入5G毫米波时变信道研究领域，研究其时变特性，建立路损和信道参数的时变模型。同时开展毫米波时变信道仿真技术研究，修正第三代合作伙伴项目（3GPP）标准推荐的5G信道仿真平台，利用时变信道模型与参数实现信道精准仿真，以再现5G毫米波时变信道传播特性。

The channel models for third- and fourth-generation (3G and 4G) mobile radio systems are geometry based stochastic models, because of their empirical path-loss models and statistical channel parameters, every now and then big deviations from the actual channels are found especially for the link level simulation and design. Millimeter wave (mmWave) transmission is a key technology in 5G radio access and backhaul, time-varying channel modeling is one of the hot topics and research directions due to large amount of moving scattereres existed in 5G tyipcal scenarios. MmWave channel measurements will be carried out in this project at ITU mmWave frequency candidates for 5G , the artificial neural networks (ANN) and stochastic differential equation (SDE) are introduced to study the time varying channel properties and further to develop its corresponding models for path-loss and channel parameters. Meanwhile, the time-varying channel simulation technologies will be investigated and 3GPP recommended 5G simulation platform will be modified by using the time-varying channel models and parameters to have accurate channel simulations and to playback the characteristics of 5G mmWave time-varying channels.