Robust capacity maximization transceiver design for MIMO OFDM systems

Abstract

In this paper, we investigated capacity maximization problem for Multiple Input Multiple Output Orthogonal Frequency Division Multiplexing systems with imperfect channel state information (CSI). To the best of our knowledge, the considered problem is still an open problem. However, the transceiver designs for MIMO OFDM systems have been extensively studied. It seems nobody gives closed-form solutions for resource allocation for MIMO OFDM systems with statistical channel estimation errors up to date. In our work, based on practical channel estimation algorithm, the channel estimation errors are first derived and then the robust resource allocation problem has been formulated. The structure of the optimal robust precoder is first derived, based on which the optimization problem will be simplified significantly. Furthermore, based on the Lagrangian dual method, a robust power allocation algorithm is proposed. The proposed power allocation can be considered as a variant of water-filling solution named cluster water-filling solution. Finally, simulation results show that our proposed robust design outperforms the non-robust design in terms of channel capacity.

创新点

本文中,我们研究了在不完美信道状态信息下,多天线正交频分复用系统的容量最大化问题。这个问题仍然是一个开放性的课题,然而,我们进一步研究了多天线正交频分复用系统中的收发机设计问题。到目前为止,还没有相关的文献给出统计信道估计误差情况下,多天线正交频分复用系统的资源分配的闭式解。本文中,基于实际的信道估计算法,我们推导出了信道估计误差以及给出了资源分配问题的优化形式。我们首先给出了鲁棒性预编码的结构,,在此基础上,优化问题可以进一步被简化。另外,根据拉格朗日对偶算法,我们提出了一种鲁棒性的功率分配算法。提出的功率分配算法可以看做注水算法的一种变形,称为分簇注水算法。最后,仿真结果证明了我们提出的鲁棒性设计要优于非鲁棒性的设计。