系统工程与电子技术 ›› 2018, Vol. 40 ›› Issue (7): 1615-1625.doi: 10.3969/j.issn.1001-506X.2018.07.29

• 通信与网络 • 上一篇    下一篇

多小区全双工MIMO系统中干扰对齐方案

景小荣1,2, 张翠翠1   

  1. 1. 重庆邮电大学通信与信息工程学院, 重庆 400065;
    2. 移动通信技术重庆市重点实验室, 重庆 400065
  • 出版日期:2018-06-26 发布日期:2018-06-28

Interference alignment scheme for multi-cell full-duplex MIMO systems

JING Xiaorong1,2, ZHANG Cuicui1   

  1. 1. School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications,
    Chongqing 400065, China; 2. Chongqing Key Lab of Mobile Communications Technology, Chongqing 400065, China
  • Online:2018-06-26 Published:2018-06-28

摘要:

同时同频全双工(full-duplex, FD)与多输入多输出(multiple-input multiple-output,MIMO)技术的结合尽管可成倍地提升MIMO系统的频谱效率,但由此而引起更为复杂的干扰问题限制了FD技术的实际应用。为此,基于干扰对齐(interference alignment,IA)思想,提出一种适用于多小区FD-MIMO系统的干扰解决方案。该方案同时适用于FD基站FD用户(FD-BS-FD-User)和FD基站半双工(half-duplex,HD)用户(FD-BS-HD-User)两种模型。该方案首先根据多对一(many-to-one,MTO)线性IA算法,通过闭式求解来消除小区间和小区内上行干扰及下行干扰;然后基于上行用户发送对下行用户接收所造成的残留干扰,构建目标函数,接着结合一辅助函数将该目标函数转化为仅与上行用户内层预编码矩阵相关的复合代价函数,进而基于格拉斯曼流形上的最速梯度下降法在上行用户端实现预编码矩阵的迭代求解,以抑制上行用户发送对下行用户接收所造成的干扰。数值仿真结果验证了该IA方案在多小区FD-MIMO系统中的适用性和有效性。

Abstract:

The combination of co-time co-frequency full-duplex (FD) and multiple-input multiple-output (MIMO) techniques has the potential to double the spectral efficiency of the MIMO systems, but more complicated interferences arise, which limits the application of FD technique. Based on the concept of the interference alignment (IA), this paper presents a scheme for the multi-cell FD-MIMO systems to solve the problem. This scheme is both suitable for the FD base FD user (FD-BS-FD-User) mode and FD base half duplex user (FD-BS-HD-User) mode. In the scheme, the many to one (MTO) linear IA algorithm is proposed to eliminate the inter-cell and intra-cell interference of uplink and downlink in the closed form. Then the objective function is constructed based on the residual interference between uplink users transmitting and downlink users receiving(I-UuT-DuR),and then transformed into the composite cost function with the aid of the auxiliary function,which is only related to the uplink precoding matrix. Finally, the composite cost function is optimized to obtain the precoder with the gradient descent method on the Grassmann manifold in an iterative manner. Numerical simulation results verify the applicability and effectiveness of the proposed scheme for multi-cell FD-MIMO systems.