摘要
语义通信的目的在于传输信源数据中的语义信息,通过语义提取和压缩可大幅减少网络中需要传输的数据量,降低带宽消耗和传输时延,同时在提高传输可靠性上展现出了巨大潜力。传统通信中资源分配方案都是以优化比特传输速率而设计的,并不适用于聚焦语义信息传输的语义通信网络。为利用语义通信系统优势,需要从语义层面考虑资源分配方案的设计,以进一步提高信息传输效率。首先梳理并总结了目前语义感知通信网络中资源分配技术的研究进展,然后通过分析语义感知资源分配面临的挑战,提出了一种基于任务卸载的多维资源联合优化架构,最后以面向文本的语义任务为例,给出了两种语义感知资源分配方案,通过仿真证明了方案的有效性。
Semantic communication aims to transmit the semantic information of the source data.By semantic extraction and compression,the transmission overhead can be reduced significantly,thus reducing the bandwidth consumption and the transmission latency,while showing a great potential in improving transmission reliability.The resource allocation schemes in traditional communication are designed to optimize the bit-based data rate,which is not applicable for semantic information tranmission in semantic communication networks.For utilizing the advantage of the semantic communication system,it is necessary to consider the design of the resource allocation scheme in the semantic domain to further improve the information transmission efficiency.First,the current research progress of resource allocation techniques in semantic-aware communication networks are reviewed and summarized.Via analyzing the challenges of semantic-aware resource allocation,a multi-dimensional resource joint optimization architecture is proposed based on task offloading.Finally,taking text-oriented semantic tasks as an example,two semantic-aware resource allocation schemes are provided,and the effectiveness of the scheme is verified by simulation.
作者
秦志金
冀泽霖
严蕾
陶晓明
QIN Zhijn;JI Zelin;YAN Lei;TAO Xiaoming(Department of Electronic Engineering,Tsinghua University,Beijing 100084,China;School of Electronic Engineering and Computer Science,Queen Mary University of London,London E14NS,UK;School of Telecommunications Engineering,Xidian University,Xi'an 710071,China)
出处
《移动通信》
2023年第4期25-30,共6页
Mobile Communications
基金
国家自然科学基金项目(62293484,61925105)。
关键词
语义通信
资源分配
深度学习
semantic communication
resource allocation
deep learning
