首页|Findings on Artificial Intelligence Reported by Investigators at School of Elect rical and Information Engineering (Artificial Intelligence-assisted Accurate Spe ctrum Prediction In Design of Terahertz Fiber Operating In 6g Communication Wind ow)
Findings on Artificial Intelligence Reported by Investigators at School of Elect rical and Information Engineering (Artificial Intelligence-assisted Accurate Spe ctrum Prediction In Design of Terahertz Fiber Operating In 6g Communication Wind ow)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning DailyNews Daily News-New research on Artificial Intelligenc e is the subject of a report. According tonews originating from Tianjin, People 's Republic of China, by NewsRx correspondents, research stated,"Accurate spect rum prediction in design of terahertz (THz) devices remains challenging, especia lly forTHz fibers. In this article, we propose an approach based on artificial intelligence (AI) assisted finiteelement method (FEM) to achieve accurate spect rum prediction for the design of THz fiber operating in6G communication window. "Financial support for this research came from National Natural Science Foundatio n of China (NSFC).Our news journalists obtained a quote from the research from the School of Elect rical and InformationEngineering, "The antiresonant THz fiber has been selected to verify the effectiveness of this method.Initially, the principle and physic al modeling of antiresonant THz fiber are analyzed. The spectra of THzfibers wi th different structural parameters are designed and predicted by FEM simulation. Then, the THzfibers are fabricated by 3D printing technology and the spectra a re measured by a THz time domainspectroscopy system. The spectral dataset obtai ned by both forms are prepared for the modeling ofAI-assisted FEM methods. Diff erent AI algorithms are induced in FEM to predict experimental spectrum,includi ng elman neural network (Elman), support vector machines (SVM), and general regr ession neutralnetwork (GRNN). The prediction performance obtained by different methods are compared and analyzedcomprehensively to confirm the effectiveness o f proposed methods. The AI-assisted FEM methods showgreat improvement of predic tion accuracy in the design of THz fibers."
TianjinPeople's Republic of ChinaAsi aArtificial IntelligenceEmerging TechnologiesMachine LearningSchool of E lectrical and Information Engineering
NETL
NSTL
