J4 ›› 2013, Vol. 40 ›› Issue (4): 167-173+196.doi: 10.3969/j.issn.1001-2400.2013.04.028

• 研究论文 • 上一篇    下一篇

SIMO系统中4PAM/16QAM信号的半正定规划盲序列检测算法

薛江;彭华;马金金   

  1. (解放军信息工程大学 信息工程学院,河南 郑州  450002)
  • 收稿日期:2012-03-27 出版日期:2013-08-20 发布日期:2013-10-10
  • 通讯作者: 薛江
  • 作者简介:薛江(1986-),男,解放军信息工程大学硕士研究生,E-mail:xuejiang0359@163.com.
  • 基金资助:

    国家自然科学基金资助项目(61072046)

Blind semi-definite programming sequence detection of the 4PAM/16QAM signal in the SIMO system

XUE Jiang;PENG Hua;MA Jinjin   

  1.  (School of Information Engineering, PLA Information Engineering Univ., Zhengzhou  450002, China)
  • Received:2012-03-27 Online:2013-08-20 Published:2013-10-10
  • Contact: XUE Jiang

摘要:

针对单输入多输出(SIMO)系统,使用优化理论中的半正定规划(SDP),提出了对非常模信号4PAM/16QAM的盲序列检测算法.在SDP 盲序列检测中,首先将信号盲序列检测问题松弛为可解的SDP问题; 然后,进行SDP求解;最后,通过随机扰动将SDP解还原为原始的信号盲序列检测问题.由于基于SDP的盲序列检测问题回避了信道阶数的直接估计或隐含估计,同时SDP求解不受信道互质条件的限制,因此算法具有更广泛的适用性.仿真实验表明,该算法具有需要数据量少、不依赖信道阶数估计、不受信道互质条件限制诸优点.

关键词: 盲均衡, 盲序列检测, 单输入多输出, 半定规划, 半定松弛

Abstract:

Direct blind sequence detection of the 4PAM/16QAM signal based on semi-definite programming (SDP) for the Single-Input-Multiple-Output (SIMO) system is proposed. In the process of blind SDP detection, firstly we relax the signal blind detection problem in such a way that the relaxed SDP problem is easier to solve; then we solve the SDP problem; thirdly we use randomization strategy to convert the SDP solution to the original blind signal detection problem. The SDP blind detection problem does not depend explicitly or implicitly on the channel order, nor needs the assumption that the parallel channels should be coprime, so it has extensive applicability. Simulation shows that the proposed algorithm demonstrates the advantage of a fast convergence rate and complete independence of channel order estimation and coprime assumption.

Key words: blind equalization, second-order statistics, single input multiple output, infinite impulse response, common zeros, minimum mean square error

中图分类号: 

  • TN911.23