基于MindSpore框架的智能通信实验教学  

Experimental Teaching of Intelligent Communications Based on MindSpore Framework

作　　者：袁磊[1] 杨伟明 郭宇航 YUAN Lei;YANG Weiming;GUO Yuhang(College of Information Science and Engineering,Lanzhou University,Lanzhou 730030,China)

机构地区：[1]兰州大学信息科学与工程学院,兰州730030

出　　处：《实验室研究与探索》2022年第6期198-202,251,共6页Research and Exploration In Laboratory

基　　金：甘肃省自然科学基金项目(18JR3RA268)。

摘　　要：为了适应新工科建设和人工智能在未来无线通信系统设计中应用的需求,兰州大学通信工程专业借助华为公司自研的人工智能框架MindSpore,将最新智能通信领域研究成果引入教学中,设计了基于深度学习的数字调制类型盲识别和LDPC码译码实验。该实验有助于通信工程专业学生深刻理解深度学习在设计智能通信系统的应用,为培养适应人工智能时代的通信工程师做了有益尝试。For meeting the needs of emerging engineering and application of artificial intelligence in the design of future wireless communication systems,with the help of the artificial intelligence framework MindSpore developed by Huawei enterprise,Lanzhou University introduces the latest research on intelligent communications into teaching,and designs the experiments,which are based on deep learning,on automatic modulation classification and low density parity check(LDPC)decoding algorithm.These experiments give students majoring in communication engineering a deep understanding of how deep learning can be used to design intelligent communication systems,and have made useful attempts to train communication engineers adapted to the age of artificial intelligence.

关 键 词：智能通信 MindSpore框架 实验教学 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]