您好, 访客   登录/注册

基于GMD的卫星MIMO系统预编码技术

来源:用户上传      作者:

  摘要:将全频率多波束卫星系统建模为MIMO系统,分析其信道特征,在此基础上进行预编码设计以提高系统容量。针对SLNR/SSLNR预编码算法BER性能损失大的问题,基于GMD矩阵分解对SSLNR算法进行改进,结合其信道特点减小预处理矩阵的维度,提出GMD-SSLNR算法。首先基于用户间距进行分组,每组分别对等效信道矩阵进行GMD运算,能有效降低码流间的增益差,并采用THP方法在发端抵消掉已知的组间干扰泄漏。分析和仿真表明,相比SLNR-THP及SSLNR-THP算法,算法有效提高了系统的误码性能,并降低了运算复杂度。
  关键词:多波束卫星;MIMO通信; GMD;SSLNR;低信噪比
  中图分类号:TN927.2    文献标识码:A     文章编号:1007-9416(2020)04-0000-00
  0引言
  卫星通信系统 波束间的频率复用能有效提升系统容量[1-2]。根据文献[3][4],全频率复用多波束卫星系统可建模为分布式MIMO系统,因此可将预编码技术用于多波束卫星系统前向链路,以提高其系统容量。
  经典的基于信漏躁比[5](SLNR)和基于块对角化(BD)的预编码[6]算法均有效降低同道干扰,但在工程上是很难实现的。文献[7]的的BLR-SSLNR-THP算法中,但当用户有多个码流时,码流间的增益差较大导致系统BER性能差。为此,文献[8]提出了几何均值分解(GMD)方法,GMD方法将信道矩阵分解成对角值一致的三角矩阵,从而使得每个子信道增益相等。本文利用GMD矩阵运算,对SSLNR预编码进行改进,并充分利用卫星信道特点降低矩阵运算量,设计了基于GMD的SSLNR算法——GMD-SSLNR算法。该算法中,每组用户的不同码流具有相同的增益,能够有效提高系统的误码性能。
  1 系统模型
  2 GMD-SSLNR预编码算法
  2.1 算法原理
  按距离将用户分为三组: 则第一组用户接收的信号可表示为:
  2.2 算法复杂度分析
  广义特征值分解的计算复杂度为[9]。由于GMD-SSLNR算法在每个用户上进行广义特征值分解,所以算法总的运算复杂度为。每次GMD分解的复杂度为。因此,由于,则总的复杂度为。本文的GMD-SSLNR算法单次广义特征分解的计算复杂度为。因此其总的复杂度为。其中,表示每组的平均用户数目。而SSLNR-THP的复杂度为,GMD-SSLNR的复杂度为。可见,相比SSLNR-THP算法和SLNR算法, GMD-SSLNR算法計算复杂度较小,且随着分组数的增多和的增大而减小。   3仿真結果及分析
  本节对多波束卫星MIMO系统采用蒙特卡洛方法进行系统性能仿真,将仿真结果与SSLNR-THP算法、SLNR-THP进行对比,如图1所示。
  由图1易知,在中低信噪比区域,三种算法的误码性能相当。随着信噪比的提升,相比SLNR-THP和SSLNR-THP算法,本文算法的系统误码性能有着非常明显的优势,且随着信噪比逐渐增大,优势更加明显。在高信噪比区域,本文算法优于SSLNR-THP算法大约5.6dB,优于SLNR-THP算法大约13.2dB。这是由于本文算法通过GMD矩阵运算有效消除了各用户之间的增益差。
  4 结语
  本文提出一种适用于卫星MIMO系统的GMD-SSLNR算法。算法首先利用SSLNR求得等效矩阵,运算过程中充分利用信道特点,降低矩阵维度,然后对其进行GMD分解。复杂度分析表明,相比SLNR-THP及SSLNR-THP算法,本文算法运算复杂度相对较低。仿真结果表明,在高信噪比区域,本文算法优于SSLNR-THP算法大约5.6dB,优于SLNR-THP算法大约13.2 dB。说明本文算法通过GMD矩阵运算有效消除了各用户之间的增益差。因此,本文算法有一定的工程应用价值。
  参考文献
  [1] Vidal O, Verelst G, Lacan J, et al. Next generation high throughput satellite system[C]// Proceedings of IEEE First AESS European Conference on Satellite Telecommunications. Rome:IEEE,2012:1-7.
  [2] Arapoglou P, Liolis K, Bertinelli M, et al. MIMO over satellite: a review [J].IEEE Communications Surveys & Tutorials,2011,13(1):27-51.
  [3] 王杨,赵旦峰,廖希.多波束卫星系统中低复杂度分组预编码算法[J].哈尔滨工业大学学报,2015,47(3):77-82.
  WANG Yang,ZHAO Danfeng,LIAO Xi. Low-complexity group precoding in multi-beam satellite systems [J].Journal of Harbin Institute of Technology,2015,47(3):77-82.
  [4] CHEN C, CHO T, CHUNG W. Block-lattice-reduction-aided Tomlinson-Harashima precoder designs for MU-MIMO downlink communications with clusters of correlated Users [J].IEEE Transactions on Vehicular Technology,2014,63(3):1146-1159.
  [5] C P, T M, Z W. A New SLNR-Based Linear Precoding for Down-link Multi-User Multi-Stream MIMO Systems[J].IEEE Communications Letters,2010,14(11):1008-1010.
  [6] Spencer Q H,Swindlehurst A L,Haardt M. Zero-forcing methods for downlink spatial multiplexing in multi-user MIMO channels[J].IEEE Transactions on Signal Processing,2004,52(2):461-471.
  [7] S M, B H. Solution of the multiuser downlink beamforming problem with individual SINR constraints[J].IEEE Transactions on Vehicular Technology,2004,53(1):18-28.
  [8] Jiang Y, L J, Hager WW. Joint transceiver design for MIMO communications using geometric mean decomposition[J]. IEEE Trans. Signal Process.,2005,53(10):3791 -3803.
  [9] 程云鹏,张凯院,徐仲.矩阵论[M].西安:西北工业大学出版社,2008.
  Cheng Yunpeng,Zhang Kaiyuan,Xu Zhong.Matrix Theory[M].Xian:anaorthwestern Polytechnic University Press,2008.
  收稿日期:2020-02-22
  作者简介:陶凯(1987—),男,山东潍坊人,博士,工程师,研究方向:无线通信。
  Precoding Technology for Satellite MIMO System Based on GMD
  TAO Kai1,CHEN  Qiang2
  (1.The 54th Research Institute of CETC, Shijiazhuang Hebei 050081;   2.Unit 73676, PLA, Wuxi Jiangsu 214400)
  Abstract: Multi-beam satellite with full frequency multiplexing is modeled to MIMO system, and then its channel character is analyzed, based on which the precoding algorithm is designed to improve the system throughput. The existing SLNR/SSLNR precoding algorithms are all with loss of BER. In this paper, the SSLNR algorithm is improved based on GMD matrix decomposition, and the dimension of the preprocessing matrix is decreased because of the channel character and a new precoding algorithm named GMD-SSLNR is proposed The users are grouped based on the distance between them and each group decomposes the equivalent channel matrix respectively, which can reduce the gain difference among the data streams, and the known leaking interference of the transmitter is offset with the THP algorithm. Analysis and Simulation results show that compared to SLNR-THP and SSLNR-THP, the proposed algorithm can improve the BER performance and reduce the complexity.
  Key words: Multi-beam Satellite; MIMO Communications; GMD; SSLNR; Lower BER
转载注明来源:https://www.xzbu.com/8/view-15251760.htm