K用户干扰信道的干扰对准及自由度分析
摘 要
关于干扰对齐，一直备受争议，是一个较为激进的论点。在相对比较短的一段时间里，这一概念挑战了对于有线和无线网络吞吐量限制的传统知识。一个典型的例子，有个发射机和接收机的无线K用户干扰信道，在此，由于干扰对齐，每个用户的传输速率可以为无干扰时信道容量的一半，给它期望的接收端发送信息。尽管这里的可以无限大，这就表明K用户干扰信道基本上是不受限制。然而迄今为止，干扰对齐的显著成效表明大部分都过于理想化的假设，比如，信道全部信息，带宽扩展，无线分辨率，高信号强度和显著延迟。本文先分析了干扰对齐的基本原理，自由度表征的信道容量，以及典型的X信道和高斯干扰信道。分析迭代干扰对齐算法，早前的关于干扰对齐方案一个很大的顾虑在于获取全部信道信息。然而在本文中，我们提供的迭代算法的示例，利用无线网络的对偶信道，在只知道本地信道信息的情况下实现干扰对齐，本文研究的算法，都是基于理想状态下的信道状态，且都是可逆信道，着重研究的两种分布式算法，即最大信干噪比算法和最小干扰泄漏算法，这二者的共同点都是用迭代算法寻求较为可靠地预编码矢量和干扰抑制矩阵。相对不同的地方在于，前者目的是最大化自己接收机的信干噪比，后者主要最小化接收端泄露的干扰功率，都能有效地减少干扰，达到干扰对齐。经过实验仿真结果分析，得到了一些和传输速率曲线，也对干扰对齐的可行性条件给出了一些见解。
关键词 干扰对齐 自由度 分布式处理 迭代算法 MIMO
ABSTRACT
There has been a lot of controversy on interference alignment, which is a more radical argument. In a relatively short period of time, this concept challenges the traditional knowledge of throughput constraints for wired and wireless networks. A typical example is a wireless K user interference channel with transmitter and receiver. Here, because of interference alignment, the transmission rate of each user can be half of the capacity of the channel without interference, and send information to the expected receiving terminal. Although it can be infinitely large, this shows that the K user interference channel is basically unrestricted. However, until now, the effectiveness of interference alignment has shown that most of the assumptions are too idealized, for example, all channel information, bandwidth expansion, wireless resolution, high signal strength and significant delay. In this paper, the basic principle of interference alignment, the channel capacity of the degree of freedom, and the typical X channel and Gauss interference channel are analyzed. An analysis of the iterative interference alignment algorithm, one of the early concerns about the interference alignment scheme is to obtain all channel information. However, in this paper, we provide examples of iterative algorithm, using the dual channel wireless network, only know the local channel information under the condition of interference alignment, the algorithm is based on channel state under ideal condition, and are reversible channel, two kinds of distributed algorithm research, i.e. the maximum signal noise ratio algorithm and algorithm stem leakage minimum interference, these two points is the iterative algorithms to seek reliable pre encoding vector and matrix interference suppression. The relative difference is that the former aims to maximize the signal to noise ratio of the receiver, and the latter minimizes the interference power at the receiver, which can effectively reduce interference and achieve interference alignment. Through the analysis of experimental simulation results, some and transmission rate curves are obtained, and some opinions on the feasibility of interference alignment are also given.
Key word interference align degree of freedom distributed processing iterative algorithm MIMO
目 录
符 号 表 4
缩略语表 5
第1章 绪论 1
1.1 干扰对齐的背景研究 1
1.2 干扰对齐的研究现状 2
1.3 论文结构 3
第2章 干扰对齐基本概念 4
2.1 信道信息 4
2.2 自由度 4
2.3 干扰对齐的思想起源 6
2.4 对齐原理 9
2.5 干扰对齐关键技术 10
2.6 本章小结 11
第3章 干扰对齐算法的研究 12
3.1 X信道 12
3.2 K用户高斯干扰信道 14
3.3 干扰对齐算法分类 16
3.4 空间干扰对齐算法 16
3.5 时域干扰对齐算法 20
3.6 频域干扰对齐算法 21
3.7 仿真结果 22
3.8 本章小结 24
第4章 干扰对齐的可行性 25
4.1 线性干扰解决方案 25
4.2 合适与不合适系统 26
4.3 本章小结 28
第5章 干扰对齐的回顾与展望 29
5.1 干扰对齐的回顾 29
5.2 干扰对齐的展望 30
结 论 31
论文工作总结 31
未来工作展望 31
致 谢 32
参 考 文 献 33
