一种地空3D-Massive MIMO信道模型

A 3D-Massive MIMO ground-air channel model

为了研究地空3D-Massive MIMO系统信道特性,提出一种基于三维单跳同心圆环散射体的地空3D-Massive MIMO信道模型。相比传统地空3D MIMO信道模型有两方面改进,首先是建立了基于单跳圆环散射体3D几何精确位置关系的球面波天线阵元相位偏移模型,其次采用生灭过程对Massive MIMO阵列上的非平稳特性进行建模。推导了该模型的信道统计特征函数,通过对特征函数理论推导值与蒙特卡洛统计值进行仿真对比,验证了理论推导正确性。此外,对所提信道模型与基于平面波的传统地空3D MIMO信道模型的空间相关性进行了分析及仿真对比,结果表明:在远场环境下,两者基本一致;在近场环境下,传统平面波模型不再适用,两者存在差异。最后,对若干组表征信道特征的仿真结果与公开文献中地空信道实际测试数据进行对比分析。表明本模型作为普遍适用性模型,可以根据信道的测试条件、测试环境和测试结果来优化参数配置以趋近于实际信道,以对特定信道进行精确建模。

In order to analyze the channel characteristic of ground-air 3D-Massive MIMO system,this paper has presented a ground-air 3D-Massive MIMO channel model of geometrically-based single-bounce concentric-cylinders scattering. Compared with traditional ground-air 3D MIMO channel model, the proposed model has two improvements. First, a model for phase shift of antenna array element of spherical wave is established based on the 3D precise geometric relationship of the single- bounce concentric-cylinders scattering. Second, a birth-and-death model for the non-stationary characteristics in the massive MIMO array direction is developed. The channel statistical characteristic function of the proposed model is derived, and the theoretical deduction is verified by comparing the results of the characteristic function theory value and Monte Carlo statistics value. In addition, the differences of spatial correlation between the proposed model and the traditional ground-air 3D MIMO channel model based on plane wave are analyzed and compared by simulation. The simulation results show that the two models have almost no difference under the near field. And there are differences between the two models under the far field, the traditional model is no longer applicable. Finally, the simulation results of several channel characterizations and the actu- al test data of open literatures for ground-air channel are compared and analyzed. The results show that, as a general appli- cability model, the parameters of proposed model can be optimized according to the test conditions, test environment and test results of the channel in order to approach the actual channel to accurately model the specific channel.