• h1_272145
  • 76.7KB
  • zip
  • 0
  • VIP专享
  • 0
  • 2022-05-31 08:01
matlab最简单的代码MIMO中的AMP 用于大规模MIMO检测的近似消息传递(AMP) 文件描述 AMPG.m:在先验符号上使用高斯分布的AMP算法。 AMPT.m:在先前符号上使用{0,1,-1}的AMP算法。 main_AMPG_LMMSE.m:独立的主文件,用于绘制AMPG和线性MMSE(LMMSE)的符号错误率(SER)性能。 main_massive_detection.m:用于将混合解码与LMMSE和AMPG的SER性能进行比较的主文件,其中可以通过使用LMMSE + AMPG或LMMSE + AMPT来执行混合解码。 版权 如果您在研究中使用我们的代码,请引用以下论文以感谢我们的工作: 吕善刚,丛玲:混合向量扰动预编码:近似消息传递的祝福。 IEEE Trans。 信号处理。 67(1):178-193(2019)。 维护者 珊香柳 暨南大学副教授,广州 电子邮件: 主页: 历史 这些文件已发布在我的个人网站(版本1)和MATHWORK(版本2)上。 当我在2014年左右开始攻读博士学位时,我的一项研究项目是应用AMP算法解决晶格的最近向量问题(CVP),而流行的MI
  • AMP-in-MIMO-main
  • FIG1-SER.jpg
  • AMPT.m
  • main_AMPG_LMMSE.m
  • FIG2-SER.jpg
  • main_massive_detection.m
  • AMPG.m
# AMP-in-MIMO Approximate Message Passing (AMP) for Massive MIMO Detection # File Description AMPG.m: the AMP algorithm using Gaussian distributions on the prior symbols. AMPT.m: the AMP algorithm using {0,1,-1} on the prior symbols. main_AMPG_LMMSE.m: the stand-alone main file that plots the symbol error rate (SER) performance of AMPG and linear MMSE (LMMSE). main_massive_detection.m: the main file that compares the SER performance of hybrid decoding with LMMSE and AMPG, where hybrid decoding can be performed by either using LMMSE+AMPG or LMMSE+AMPT. ![main_AMPG_LMMSE](FIG1-SER.jpg) ![main_massive_detection](FIG2-SER.jpg) # Copyright If you use our codes in your research, please acknowledge our work by citing the following paper: Shanxiang Lyu, Cong Ling: Hybrid Vector Perturbation Precoding: The Blessing of Approximate Message Passing. IEEE Trans. Signal Process. 67(1): 178-193 (2019). # Maintainer Shanxiang Lyu Associate Professor, Jinan University, Guangzhou Email: Homepage: # History These files were posted on my personal website (, edition-1) and MATHWORK (, edition-2). When I started my PhD in around 2014, one of my research projects is to apply the AMP algorithm to solve the closest vector problem (CVP) of lattices, and the popular MIMO detection problem is no more than a special case of CVP. The initial attempt was to understand the technical paper that I noted as BM11 ("The dynamics of message passing on dense graphs, with applications to compressed sensing", Mohsen Bayati AND Andrea Montanari, IEEE TIT, 2011). Later I tried to simplify Belief Propagation to derive the set of equations used in AMP (see Sec. IV in "Hybrid Vector Perturbation Precoding: The Blessing of Approximate Message Passing", S. Lyu and C. Ling, IEEE TSP, 2019). Regardless of the versions of AMP that we are using, a critical step in its MIMO detection application is to assign the a priori distribution to the "x" of "y=H*x+n". While in the transmission "x" is admitting a uniform distribution over QAM symbols, in detection we may temporarily assume that "x" admits a Gaussian distribution, and the estimated "x" can be further quantized to obtain discrete QAM symbols. I call the AMP algorithm using this Gaussian prior as AMPG, and this algorithm is surprisingly simple with only 3 lines of codes. While the AMPG algorithm is slightly different with those in BM11, the principles are the same. In addition, Jeon and Studer published their result of using the exact priori in AMP (AMP-exact) in the 2015 ISIT, which certainly outperforms AMPG in the SER performance, so I decided to post the codes (edition-1) on my personal website in 2015. When it comes to 2017, to address the issue that AMP-exact cannot perform well when the range of “x” is large (beyond +-1 and small QAM), I developed a scheme that aims to bypass this constraint. Specially, in a two-step procedure, we first adopt a low-complexity algorithm to temporarily estimate “x”. If this estimate is not too far from the actual “x”, we can employ AMP with a smaller range of prior symbols. The simplest case is to assign only “0,1,-1” to “x”, and this algorithm is noted as AMPT. While this hybrid decoding algorithm can perform well in MIMO detection, the MIMO precoding problem seems technically more general, so I decided to formulate the stuff in the precoding context and submitted it to IEEE TSP. Nevertheless, the application to MIMO detection is still formulated in Section VII of the IEEE TSP 2019 paper.
    • 电力系统状态估计MATLAB算法
      状态估计算法 MATLAB 内附readme 详细说明了使用方法和步骤 有专门的txt文件 可以输入自己的bus阵 line阵等 即可进行状态估计
    • matlab算法经典程序
    • 粒子滤波matlab算法
    • matlab算法大全
    • Matlab算法大全
      Matlab算法 PDF 分章阅读 高清PDF 很好很全面的阅读材料
    • 各种MATLAB算法
    • Matlab算法大全
      Matlab算法大全 第01章线性规划 第02章整数规划 第03章非线性规划 第04章动态规划 第05章图与网络 第06章排队论 第07章对策论 第08章层次分析法 第09章插值与拟合 第10章数据的统计描述和分析 第11章方差分析 第12章...
    • matlab算法大全
    • matlab算法大全
      matlab 中常用的程序 函数使用示例
    • matlab 算法程序
      matlab 算法程序,包括了插值、函数逼近、数值积分、非线性方程求解、统计分析、偏微分方程数值解法等17个部分,每个部分针对各种函数有m文件代码和相关解释说明。