J4 ›› 2013, Vol. 40 ›› Issue (6): 25-31+84.doi: 10.3969/j.issn.1001-2400.2013.06.005

• 研究论文 • 上一篇    下一篇

低功耗并行MIMO空间复用检测器的VLSI结构

李晓峰1;冯大政1;胡树楷2
  

  1. (1. 西安电子科技大学 雷达信号处理国家重点实验室,陕西 西安  710071;
    2. 西安电子科技大学 综合业务网理论及关键技术国家重点实验室,陕西 西安  710071)
  • 收稿日期:2012-09-04 出版日期:2013-12-20 发布日期:2014-01-10
  • 通讯作者: 李晓峰
  • 作者简介:李晓峰(1979-),男,西安电子科技大学博士研究生,E-mail: superbigbignose@gmail.com.
  • 基金资助:

    国家自然科学基金资助项目(60971111);国家重点基础研究发展计划(973)资助项目(2010CB328300)

Low power parallel VLSI architecture for  spatially multiplexed MIMO detection

LI Xiaofeng1;FENG Dazheng1;HU Shukai2   

  1. (1. National Key Lab. of Radar Signal Processing, Xidian Univ., Xi'an  710071, China;
    2. State Key Lab. of Integrated Service Networks, Xidian Univ., Xi'an  710071, China)
  • Received:2012-09-04 Online:2013-12-20 Published:2014-01-10
  • Contact: LI Xiaofeng

摘要:

针对2×2天线配置的多输入多输出无线通信系统, 提出了一种新颖的近最大似然多输入多输出检测算法.此算法只对搜索树的第2层节点耗尽搜索, 第1层应用区域判定法直接得到假设星座点和反假设点,不需要对每一层搜索树的部分欧几里得距离排序, 降低了计算复杂度, 而且适合并行处理. 依据该算法设计了规则的、流水线型的超大规模集成电路, 可以配置并行检测核的数目,从而可灵活地控制吞吐率. 在40nm的工艺下,约束工作电压为1.08V,设定时钟频率为156MHz,版图实现可以达到312Mbit/s的吞吐量和23mW的功耗,处理延迟仅为0.051ns.

关键词: 球形检测, 多输入多输出空间复用, 软输出, 超大规模集成电路

Abstract:

This paper proposes a novel maximum likelihood-like detection algorithm for the 2×2 MIMO wireless communication system. Only the second layer in the search tree needs to be verified exhaustively, and then the area determination strategy is applied to the first layer to figure out the hypothesis node and counter-hypothesis nodes without sorting the partial Euclidean distance for each layer, which reduces the complexity significantly and leads to parallel implementation. The regular VLSI architecture is designed according to the detection algorithm, and flexible parallelism can well control the throughput for versatile applications. The VLSI architecture is implemented with 40nm technology and constrained with the voltage of 1.08V and the clock frequency of 156MHz. The detector achieves 312Mbit/s throughput with the power dissipation of 23mW and the latency of 0.051ns .

Key words: sphere detection, spatially multiplexed multiple-input multiple-output, soft output, very large scale integration