格约减理论在MIMO系统中的应用

项目名称： 格约减理论在MIMO系统中的应用

项目编号： No.61201198

项目类型： 青年科学基金项目

立项/批准年度： 2013

项目学科： 电子学与信息系统

项目作者： 孙艳华

作者单位： 北京工业大学

项目金额： 25万元

中文摘要： 格约减理论是研究格上著名难题最近向量问题的一个重要工具，近来虽然被成功应用在MIMO系统中，成为提高MIMO系统信号检测和预编码性能的一个有力工具，但还有很多开放问题有待解决，如目前采用的格基约减算法单一，在编码MIMO系统中的应用限制等。 鉴于此，本项目将在格基约减实现算法，信号的非线性量化，数据检测和预编码等方面展开深入的研究，探讨格理论中除目前常用的LLL和SA算法之外的其它格基约减算法，分析其复杂度；结合正交对偶格的概念提出新的约减算法及信号非线性联合量化方案；研究格约减技术在编码MIMO系统中的应用，设计软判决格约减检测算法；充分考虑信道信息部分已知时对格约减检测性能的影响；探索格约减技术在多级降秩检测中应用的可能性，提出新的基于格约减的多级降秩检测；针对不同的系统拓扑结构，研究格约减技术在有限反馈及协同中继系统预编码方面的应用，并对复杂度进行分析和比较。

中文关键词： MIMO；格约减；全双工；证据理论；

英文摘要： Lattice reduction theory is an important tool to solve the well-known closest vector problem in a lattice. Although it has been successfully applied in MIMO systems recently, and become a powerful approach to improve the performance of signal detection and precoding of MIMO systems. However, there are a number of open questions to be resolved such as lattice basis reduction algorithm is limited to LLL and SA algorithms to date, the application limit in the coded MIMO systems and so on. In view of this, the project will focus on the following aspects: lattice basis reduction algorithms, nonlinear quantization of the signal, data detection and precoding , generalize other lattice basis reduction algorithms in lattice theory in addition to LLL and SA algorithm, and propose a new reduction algorithm combined with orthogonal dual lattice concept and nonlinear quantization scheme; study the application of lattice reduction in coded MIMO systems and design soft-decision lattice reduction aided detection ;explore the probability of multistage reduced-rank detection combined with lattice reduction theory and propose the new multistage reduced-rank detction based on lattice reduction; for different topology structure, research the application of lattice reduction in the precoding of the limited feedback and cooperative

英文关键词： MIMO；lattice reduction；full duplex；evidence theory；