A Receiver-centric Approach to Interference Management: Fairness and Outage Optimization
A Receiver-centric Approach to Interference Management: Fairness and Outage Optimization
复制标题
DOI：
10.1109/tit.2016.2594219
复制
发表时间：
2016-10-01
影响因子：
2.5
通讯作者：
Wang, Xiaodong
中科院分区：
文献类型：
--
作者：
Ashraphijuo, Mehdi;Tajer, Ali;Wang, Xiaodong
多用户干扰信道中的有效干扰管理需要用户协调地形成它们的传输和干扰管理决策，并使它们适应信道的状态。建立这种通常通过信息交换促进的协调在快速变化的渠道中是令人望而却步的，特别是在网络规模增长的时候。本文针对多用户高斯干扰信道，提出了一种以接收机为中心的干扰管理方法。在该方法中，发射机根据独立的高斯码本采用速率分割和叠加编码来生成它们的消息。接收器可以以任何期望的联合或有序方式自由地解码任意组干扰消息以及它们指定的消息，并将其余的干扰者视为高斯噪声。将所提出的以接收者为中心的干扰管理方法应用于两类问题(中断优化和公平约束的速率分配)，并提供了构造性证明来建立所提出方法的下列性质：1)每个接收者要解码的最优码本集是每个接收者基于其本地信道状态信息(CSI)做出的局部决策；2)全局最优传输速率与接收者基于其本地信息计算的局部最优速率有关，这意味着发送者不需要明确的CSI知识并且可以通过来自接收者的有限反馈来确定其速率；以及3)在每个接收器处获得最优干扰管理策略具有受控的复杂性。
Effective interference management in the multiuser interference channel necessitates that the users form their transmission and interference management decisions in coordination, and adapt them to the state of the channel. Establishing such coordination, often facilitated through information exchange, is prohibitive in fast-varying channels, especially when the network size grows. This paper focuses on the multiuser Gaussian interference channel and offers a receiver-centric approach to interference management. In this approach, the transmitters deploy rate-splitting and superposition coding to generate their messages according to independent Gaussian codebooks. The receivers can freely decode any arbitrary set of interfering messages along with their designated messages in any desired joint or ordered fashion, and treat the rest of the interferers as Gaussian noise. The proposed receiver-centric interference management approach is applied to two class of problems (outage optimization and fairness-constrained rate allocation), and constructive proofs are provided to establish the following properties for the proposed approach: 1) the optimal set of codebooks to be decoded by each receiver is a local decision made by each receiver based on its local channel state information (CSI); 2) the globally optimal transmission rates are related to locally optimal rates computed by the receivers based on their local information, which implies that the transmitters do not require explicit knowledge of the CSI and can determine their rates via limited feedback from the receivers; and 3) obtaining the optimal interference management strategy at each receiver has controlled complexity.