Two-way cooperative communications with statistical channel knowledge
Two-way cooperative communications with statistical channel knowledge
复制标题
DOI：
10.1016/j.sigpro.2013.05.015
复制
发表时间：
2013-12
期刊：
影响因子：
--
通讯作者：
Fadhel A. Al-Humaidi;S. Shahbazpanahi
中科院分区：
文献类型：
--
作者：
Fadhel A. Al-Humaidi;S. Shahbazpanahi
研究了由两个收发器和多个中继节点组成的双向中继网络的分布式波束赋形问题。在这项工作中的主要假设，它不同于以前报道的结果，是假设收发器之一，只有统计信息的其他收发器和中继节点之间的信道。该假设对通过训练获得信道状态信息所需的带宽施加了不太严格的限制。基于此统计模型，我们建议使用机会约束规划方法来设计一个分布式波束形成算法。在这种方法中，我们的目标是最大限度地减少总发射功率（在整个网络中消耗）的收发器之一，受到两个概率约束。这些约束保证收发器的接收SNR的中断概率（如由主收发器感知的）不小于某些给定阈值。我们严格证明，这样的方法导致中继选择算法，其中具有最强的信道系数的主收发器的中继参与中继和其余的中继被关闭。因此，最佳分布式波束形成算法被简化为功率控制解决方案。给出了该问题的封闭解，并通过数值算例对其性能进行了评价。
We consider the problem of distributed beamforming for a two-way relay network which consists of two transceivers and multiple relay nodes. The main assumption in this work, which differentiates it from previously reported results, is that one of the transceivers is assumed to have only statistical information about channels between the other transceiver and the relay nodes. This assumption imposes less stringent restrictions on the bandwidth required to obtain channel state information via training. Based on this statistical modeling, we propose to use a chance-constrained programming approach to design a distributed beamforming algorithm. In this approach, we aim to minimize the total transmit power (consumed in the entire network) as perceived by one of the transceivers, subject to two probabilistic constraints. These constraints guarantee that the outage probability of the transceivers' received SNRs, as perceived by the master transceiver, is not less than certain given thresholds. We prove rigorously that such an approach leads to a relay selection algorithm where the relay with the strongest channel coefficient to the master transceiver participates in relaying and the remaining relays are shut off. As such, the optimal distributed beamforming algorithm is simplified to a power control solution. Closed-form solution to this problem is obtained and its performance is evaluated through numerical examples.