基于向量空间投影的多用户MIMO-CDMA系统半盲信道估计算法

发布时间：2018-03-26 04:17

本文选题：半盲信道估计　切入点：MIMO　出处：《电子科技大学》2014年硕士论文

【摘要】：在无线通信系统中,为了获得更大的性能优势,往往需要准确的估计出无线信道参数。目前常用的信道估计方法可以归纳为三类:基于训练符的信道估计方法、盲信道估计方法和半盲信道估计方法。基于训练符的信道估计方法相对简单,但是频带利用率低;盲信道估计方法频带利用率高,但是相应的计算复杂度也大;半盲信道估计方法结合了盲信道估计和非盲信道估计方法的优点,有着一定的优势。因此,频带利用率、算法复杂度以及估计精度是设计信道估计算法时需要综合考虑的三个关键问题。本文针对使用OSTBC技术的多用户MIMO-CDMA系统,提出了一种新的半盲信道估计算法。首先,本文对基于训练符的LS信道估计算法和基于MUSIC的半盲信道估计算法进行了详细的研究,分析这两种信道估计算法的优缺点,包括频带利用率和算法性能,并进行了试验仿真验证。然后,提出了一种使用OSTBC技术的多用户MIMO-CDMA系统模型,对该模型进行了详细的分析说明。针对该系统模型,在MUSIC算法的研究基础上提出了一种基于向量空间投影的半盲信道估计算法。在算法的推导过程中,证明了需要估计的信道信息向量属于两个非正交的子空间,并利用向量空间投影理论求得该向量。在具体的计算过程中,通过对正交投影矩阵的分析,结合矩阵特征值、特征向量的性质提出了一种单步解方法,减小了算法的复杂度。该算法仅使用了一组已知的训练符,便解决了估计量符号模糊问题,没有增加系统的复杂度,也没有对系统频带利用率产生影响。最后,通过仿真实验证明了本文提出的算法相比于基于训练符的LS信道估计算法有更高的估计精度和更低的误码率。本文的创新点概括如下:(1)提出了一种单步解方法,该方法将求一个维度较大矩阵的主特征向量转向求一个维度不大的矩阵主特征向量,有效减小了算法的计算复杂度;(2)通过对系统模型的研究,论证了需要估计的复合信道信息向量属于两个非正交的子空间,并使用向量空间投影的方法直接计算得到复合信道信息,避免了基于MUSIC和LS的信道估计方法中需要估计过多参数的问题;
[Abstract]:In order to gain more performance advantages in wireless communication systems, it is often necessary to estimate the wireless channel parameters accurately. The commonly used channel estimation methods can be classified into three categories: Channel estimation methods based on training characters, Blind channel estimation method and semi-blind channel estimation method. The channel estimation method based on training symbol is relatively simple, but the frequency band efficiency is low, the blind channel estimation method has high frequency band efficiency, but the corresponding computational complexity is also large. The semi-blind channel estimation method combines the advantages of the blind channel estimation method and the non-blind channel estimation method. Algorithm complexity and estimation accuracy are three key problems to be considered when designing channel estimation algorithms. In this paper, a new semi-blind channel estimation algorithm is proposed for multi-user MIMO-CDMA systems using OSTBC technology. In this paper, the LS channel estimation algorithm based on training symbol and semi-blind channel estimation algorithm based on MUSIC are studied in detail, and the advantages and disadvantages of these two channel estimation algorithms are analyzed, including frequency band efficiency and algorithm performance. Then, a multi-user MIMO-CDMA system model using OSTBC technology is proposed, and the model is analyzed in detail. Based on the research of MUSIC algorithm, a semi-blind channel estimation algorithm based on vector space projection is proposed. In the derivation of the algorithm, it is proved that the channel information vector that needs to be estimated belongs to two non-orthogonal subspaces. The vector is obtained by using the vector space projection theory. In the process of calculation, a one-step method is proposed by analyzing the orthogonal projection matrix and combining the eigenvalues of the matrix and the properties of the eigenvector. The complexity of the algorithm is reduced. The algorithm only uses a set of known training symbols, which solves the ambiguity problem of the estimator symbol, does not increase the complexity of the system, nor does it affect the frequency band efficiency of the system. Finally, The simulation results show that the proposed algorithm has higher estimation accuracy and lower bit error rate than the LS channel estimation algorithm based on training characters. The innovation of this paper is summarized as follows: 1) A one-step method is proposed. In this method, the principal eigenvector of a large dimension matrix is changed to the principal eigenvector of a small dimension matrix, which effectively reduces the computational complexity of the algorithm. It is proved that the complex channel information vector that needs to be estimated belongs to two non-orthogonal subspaces, and the compound channel information is directly calculated by vector space projection method. The problem of estimating too many parameters in channel estimation method based on MUSIC and LS is avoided.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.533

【共引文献】