| 摘要: |
| 针对传统多输入多输出（Multiple-Input Multiple-Output,MIMO）空分复用发射天线数目识别方法要求对于非合作方接收天线数目大于发射天线数目，以及特征寻找困难的问题，提出了一种基于卷积神经网络的MIMO空分复用发射天线数目识别方法。通过单天线接收目标信号，采集数据，再由卷积神经网络（Convolutional Neural Network,CNN）训练和测试。网络用发射端信号经过空时编码后表现出的不同特征识别发射天线数目。仿真和实测数据验证了CNN对于发射天线数目识别的效果，在信噪比大于15 dB时识别率可以达到85%。 |
| 关键词: MIMO 空分复用 发射天线数目识别 卷积神经网络 |
| DOI： |
|
| 基金项目:国家自然科学基金资助项目（6146105）；广西科技重大专项（桂科AA21077008）；广西高校中青年教师科研基础能力提升项目（2019KY1035）；桂林电子科技大学研究生教育创新计划项目（2020YCXS021） |
|
| MIMO space division multiplexing transmitting antennas number identification based on convolutional neural network |
| YI Guoshun,XIE Yuelei,LIANG Wenbin |
| (School of Information and Communication,Guilin University of Electronic Technology,Guilin 541004,China;1.School of Information and Communication,Guilin University of Electronic Technology,Guilin 541004,China;2.National and Local Joint Engineering Research Centre for Satellite Navigation,Positioning and Location Services,Guilin 541004,China;School of Electronic Information and Automation,Guilin Institute of Aerospace Industry,Guilin 541004,China) |
| Abstract: |
| The traditional multiple-input multiple-output(MIMO) space division multiplexing transmitting antenna number identification method requires that the number of receiving antennas of non-cooperative parties is greater than that of transmitting antennas,and it is difficult to find characteristics.To solve above problems,a method for identifying the number of MIMO space division multiplexing transmitting antennas based on convolutional neural network(CNN) is proposed.The target signal is received through a single antenna,the data is collected and then trained and tested by CNN.The network uses the different characteristics of the transmitting end signal after space-time coding to identify the number of transmitting antennas.Simulation and measured data verify the effect of the CNN on the number of transmitting antennas.When the measured data has a signal-to-noise ratio greater than 15 dB,the recognition rate can reach 85%. |
| Key words: MIMO space division multiplexing transmitting antennas number identification convolutional neural network |
