Accepted for Publication in Ieee Transactions on Signal Processing 1 Robust Qos-constrained Optimization of Downlink Multiuser Miso Systems
Accepted for Publication in Ieee Transactions on Signal Processing 1 Robust Qos-constrained Optimization of Downlink Multiuser Miso Systems
复制标题
DOI：
10.1109/tsp.2008.2008553
复制
发表时间：
2009-02
影响因子：
5.4
通讯作者：
Nikola Vuči´c;H. Boche
中科院分区：
文献类型：
--
作者：
Nikola Vuči´c;H. Boche
可以通过使用预编码技术来利用通信系统的发送侧处的信道的知识，整体传输质量可以显著地受益于该预编码技术。然而，在实际的无线系统中，信道状态信息很容易出现错误，这有时会急剧恶化的性能。在本文中，我们解决的问题，在下行链路多用户系统中的鲁棒收发器的设计，在发射机的错误的信道知识。基站配有天线阵列，而用户只有单个天线。收发器优化是在一组预定义的用户的服务质量约束下执行的，所述约束被定义为最大均方误差或最小信号与干扰加噪声比（SINR），对于属于给定的有界不确定性集合的所有干扰，必须满足所述约束。从凸优化理论出发，利用半定规划方法得到有效的数值解。所提出的算法被发现优于相关的方法在文献中所取得的性能，同时保持较低的计算复杂度。所研究的不确定性模型适用于减轻量化或信道估计的结果而出现的典型误差。
The knowledge of the channel at the transmit side of a communication system can be exploited by using precoding techniques, from which the overall transmission quality might benefit significantly. However, in practical wireless systems, the channel state information is prone to errors, which sometimes deteriorates the performance drastically. In this paper, we address the problem of robust transceiver design in a downlink multiuser system, with respect to the erroneous channel knowledge at the transmitter. The base station is equipped with an antenna array, while users have single antennas. The transceiver optimization is performed under a set of predefined users' quality-of-service constraints, defined as maximum mean square errors, or minimum signal-to-interference-plus-noise ratios (SINRs), which must be satisfied for all disturbances that belong to given, bounded uncertainty sets. Efficient numerical solutions are obtained using semidefinite programming methods from convex optimization theory. The proposed algorithms are found to outperform related approaches in the literature in terms of the achieved performance, while maintaining low computational complexity. The studied uncertainty models are applicable in mitigating typical errors that emerge as a result of quantization or channel estimation.