As Information,Communications,and Data Technology(ICDT)are deeply integrated,the research of 6G gradually rises.Meanwhile,feder-ated learning(FL)as a distributed artificial intelli-gence(AI)framework is generally believed to be the most promising solution to achieve"Native AI"in 6G.While the adoption of energy as a metric in AI and wireless networks is emerging,most studies still fo-cused on obtaining high levels of accuracy,with little consideration on new features of future networks and their possible impact on energy consumption.To ad-dress this issue,this article focuses on green concerns in FL over 6G.We first analyze and summarize major energy consumption challenges caused by technical characteristics of FL and the dynamical heterogene-ity of 6G networks,and model the energy consump-tion in FL over 6G from aspects of computation and communication.We classify and summarize the ba-sic ways to reduce energy,and present several feasible green designs for FL-based 6G network architecture from three perspectives.According to the simulation results,we provide a useful guideline to researchers that different schemes should be used to achieve the minimum energy consumption at a reasonable cost of learning accuracy for different network scenarios and service requirements in FL-based 6G network.

Borui Zhao;Qimei Cui;Shengyuan Liang;Jinli Zhai;Yanzhao Hou;Xueqing Huang;Miao Pan;Xiaofeng Tao

National Engineering Laboratory for Mobile Network Technologies,Beijing University of Posts and Telecommunications,Beijing,100876,China;Department of Computer Science,New York Institute of Technology,Old Westbury,NY,11568,USA;Department of Electrical and Computer Engineering,University of Houston,Houston,TX,77204,USA

中国通信（英文版）

[1]Borui Zhao;Qimei Cui;Shengyuan Liang;Jinli Zhai;Yanzhao Hou;Xueqing Huang;Miao Pan;Xiaofeng Tao-.Green Concerns in Federated Learning over 6G)[J].中国通信（英文版）,2022(03):50-69

feder,cused

Green,Concerns,Federated,Learning,over,6G,Information,Communications,Data,Technology,ICDT,are,deeply,integrated,gradually,rises,Meanwhile,learning,FL,distributed,artificial,intelli,gence,framework,generally,believed,most,promising,solution,achieve,Native,While,adoption,energy,metric,wireless,networks,emerging,studies,still,obtaining,high,levels,accuracy,little,consideration,new,features,future,their,possible,impact,consumption,To,dress,this,issue,article,focuses,green,concerns,We,first,analyze,summarize,major,challenges,caused,by,technical,characteristics,dynamical,heterogene,ity,model,from,aspects,computation,communication,classify,sic,ways,reduce,present,several,feasible,designs,architecture,three,perspectives,According,simulation,results,we,provide,useful,guideline,researchers,that,different,schemes,should,minimum,reasonable,cost,scenarios,service,requirements

0.595912