通信学报 ›› 2021, Vol. 42 ›› Issue (11): 97-108.doi: 10.11959/j.issn.1000-436x.2021214

• 学术论文 • 上一篇    下一篇

人工智能物联网中面向智能任务的语义通信方法

刘传宏1, 郭彩丽1,2, 杨洋2, 冯春燕1, 孙启政1, 陈九九1   

  1. 1 北京邮电大学先进信息网络北京实验室，北京 100876
    2 北京邮电大学网络体系构建与融合北京市重点实验室，北京 100876
  • 修回日期:2021-11-03 出版日期:2021-11-25 发布日期:2021-11-25
  • 作者简介:刘传宏（1998− ），男，安徽池州人，北京邮电大学博士生，主要研究方向为深度学习、语义通信、资源分配等
    郭彩丽（1977− ），女，山西太原人，博士，北京邮电大学教授、博士生导师，主要研究方向为语义通信、无线移动通信技术、认知无线电、信号检测与估值、车联网、可见光通信、视觉智能计算、社交跨媒体数据挖掘与分析等
    杨洋（1991− ），男，湖南娄底人，博士，北京邮电大学讲师，主要研究方向为可见光通信、室内定位技术、车联网技术、语义通信技术等
    冯春燕（1963− ），女，北京人，博士，北京邮电大学教授、博士生导师，主要研究方向为无线通信信息传输与处理、宽带通信网络理论与技术、社交网络分析和信息检索、电信大数据分析与挖掘等
    孙启政（1997− ），女，河南安阳人，北京邮电大学博士生，主要研究方向为语义通信、视觉内容理解、深度学习算法等
    陈九九（1994− ），男，湖南平江人，北京邮电大学博士生，主要研究方向为车联网资源分配、语义通信、强化学习算法等
  • 基金资助:
    国家自然科学基金资助项目(92067202);国家重点研发计划基金资助项目(2018YFB1800805)

Intelligent task-oriented semantic communication method in artificial intelligence of things

Chuanhong LIU1, Caili GUO1,2, Yang YANG2, Chunyan FENG1, Qizheng SUN1, Jiujiu CHEN1   

  1. 1 Beijing Laboratory of Advanced Information Networks, Beijing University of Posts and Telecommunications, Beijing 100876, China
    2 Beijing Key Laboratory of Network System Construction and Integration, Beijing University of Posts and Telecommunications, Beijing 100876, China
  • Revised:2021-11-03 Online:2021-11-25 Published:2021-11-25
  • Supported by:
    The National Natural Science Foundation of China(92067202);The National Key Research and Development Program of China(2018YFB1800805)

在线阅读

175

PDF下载

1560

摘要：

随着物联网（IoT）和人工智能（AI）技术的融合发展，传统的数据集中式云计算处理方式难以有效去除数据中大量的冗余信息，给人工智能物联网（AIoT）中智能任务低时延、高精度的需求带来挑战。针对这一挑战，基于深度学习方法提出了AIoT中面向智能任务的语义通信方法。针对图像分类任务，在IoT设备上利用卷积神经网络（CNN）提取图像的特征图；从语义概念出发，将语义概念和特征图进行关联，提取语义关系；基于语义关系实现语义压缩，减小网络传输的压力，降低智能任务的处理时延。实验和仿真结果表明，对比传统通信方案，所提方法的复杂度仅约为传统方案的0.8%，同时具有更高的分类任务性能；对比特征图全部传输的方案，所提方法传输时延降低了80%，大大提升了有效分类准确率。

关键词: 物联网, 语义通信, 图像分类, 人工智能, 语义压缩

Abstract:

With the integration and development of Internet of things (IoT) and artificial intelligence (AI) technologies, traditional data centralized cloud computing processing methods are difficult to effectively remove a large amount of redundant information in data, which brings challenges to the low-latency and high-precision requirements of intelligent tasks in the artificial intelligence of things (AIoT).In response to this challenge, a semantic communication method oriented to intelligent tasks in AIoT was proposed based on the deep learning method.For image classification tasks, convolutional neural networks (CNN) were used on IoT devices to extract image feature maps.Starting from semantic concepts, semantic concepts and feature maps were associated to extract semantic relationships.Based on the semantic relationships, semantic compression was implemented to reduce the pressure of network transmission and the processing delay of intelligent tasks.Experimental and simulation results show that, compared with traditional communication scheme, the proposed method is only about 0.8% of the traditional scheme, and at the same time it has higher classification task performance.Compared with the scheme that all feature maps are transmitted, the transmission delay of the proposed method is reduced by 80% and the effective accuracy of image classification task is greatly improved.

Key words: Internet of things, semantic communication, image classification, artificial intelligence, semantic compression

中图分类号: 

No Suggested Reading articles found!