链路特性互异下多用户OFDM/SDMA系统联合信道估计和多用户检测算法研究

项目名称： 链路特性互异下多用户OFDM/SDMA系统联合信道估计和多用户检测算法研究

项目编号： No.61271421

项目类型： 面上项目

立项/批准年度： 2013

项目学科： 无线电电子学、电信技术

项目作者： 穆晓敏

作者单位： 郑州大学

项目金额： 81万元

中文摘要： 多用户OFDM/SDMA无线通信机制有望成为未来无线通信的主流技术。本项目针对该机制下多链路、多天线MIMO信道特点，建立各链路信道统计特性各向异性（non-isotropy）的MIMO信道模型，提出多链路有效信道长度的并行搜索算法和基于期望最大化（EM）的信道估计算法；并对该模型下信道估计和多用户检测联合优化过程中的NP难解问题进行研究和探索。联合优化过程中NP难解问题的主要困难在于随着用户规模逐渐增大，运算复杂度呈指数级增长。本项目拟采用重复加权Boosting Search算法设计权重因子的自适应更新方案，提高搜索效率，降低运算复杂度，引导算法聚敛于全局最优解。本项目的研究结果将为多用户、多址、多链路场景下MIMO信道提供更切合实际的的信道模型，丰富进化算法求解信道估计和多用户检测组合优化问题的理论体系，对多用户OFDM/SDMA系统性能评价及走向实际应用提供重要的理论依据和参考。

中文关键词： 无线通信；信道估计；信号检测；正交频分复用；空分多址接入

英文摘要： As one of the predominant techniques of future wireless communications, Orthogonal Frequency-Division Multiplexing/Spatial-Division Multiple Access (OFDM/ SDMA) systems exploit the advantages of both OFDM and SDMA, which increase the systems' capacity by sharing the same bandwidths and time slots by several users roaming in different geographical. Based on the analysis of the multiple links and multiple antennae of the MIMO channel model of the considered multi-user systems. We will construct a statistically Non-isotropy MIMO channel model and then we will study the parallel search algorithm of the effective channel lengths for multiple channel links scenario and research on the joint channel impulse response and noise-variance estimation scheme based on expectation maximization algorithm as well as its convergence ability and convergece rates. What is more, we will explore the Non-deterministic Polynomial (NP) hard problems of joint channel estimation and multi-user detection. The challenge of the joint channel estimation and multi-user detection in multi-user systems is that the computational complexity is exponentially increasing as the number of users supported by the system. Furthermore, we will design an adaptive updating mechanism to elevate the search efficiency and reduce the computational complexity. T

英文关键词： Wireless communication；channel estimation；signal detection；Orthogonal Frequency-Division Multiplexing；Spatial-Division Multiple Access