| 摘要: |
| 针对大规模多输入多输出(Multiple-Input Multiple-Output,MIMO)系统中近似最优线性最小均方误差(Minimum Mean Square Error,MMSE)算法复杂度过高问题，提出了RC-CG(Region Constellation-Conjugate Gradient)低复杂度近似最优信号检测算法。该算法首先利用共轭梯度(Conjugate Gradient,CG)迭代算法避免MMSE信号检测算法的高维度矩阵求逆，降低计算复杂度；其次引入二分查找算法对星座图进行区域分块，优化迭代初始解，使算法在保证原来检测性能的基础上加快收敛速度。仿真结果表明，该算法不仅可以达到近似MMSE算法的检测性能，而且适用于高阶调制，算法复杂度从O(K3)降低到O(K2)。 |
| 关键词: 大规模MIMO 信号检测 CG迭代算法 二分查找算法 |
| DOI： |
|
| 基金项目:重庆市重点产业共性关键技术创新重大主题专项项目(cstc2017zdcy-zdzx0030) |
|
| An Improved CG Detection Algorithm for Massive MIMO Systems |
| LI Chen,DENG Bingguang,ZHANG Zhizhong |
| (School of Communication and Information Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,China) |
| Abstract: |
| For the high complexity problem of approximate optimal linear minimum mean square error(MMSE) algorithm in massive multiple-input multiple-output(MIMO) systems,a region constellation-conjugate gradient(RC-CG) approximate optimal signal detection algorithm is proposed.Firstly,the CG iterative algorithm is used to avoid the high dimensional matrix inversion of MMSE signal detection algorithm and reduce the computational complexity.Then,a binary search algorithm is proposed to regionally block the constellation and optimize the iterative initial solution by simply checking the values of received signals,which can accelerate the convergence rate of the algorithm while guaranteeing the original detection performance to reduce the complexity further.Simulation results indicate that the proposed algorithm not only achieves the near-optimal performance of MMSE algorithm but also applies to high-order modulation.The algorithm complexity is reduced from O(K3) to O(K2). |
| Key words: massive MIMO signal detection CG iterative algorithm binary search algorithm |
《电讯技术》编辑部 