Elisa Zimaglia, D. Riviello, R. Garello, R. Fantini
{"title":"A Novel Deep Learning Approach to CSI Feedback Reporting for NR 5G Cellular Systems","authors":"Elisa Zimaglia, D. Riviello, R. Garello, R. Fantini","doi":"10.1109/MTTW51045.2020.9245055","DOIUrl":null,"url":null,"abstract":"In this paper, we study 5G Channel State Information feedback reporting. We show that a Deep Learning approach based on Convolutional Neural Networks can be used to learn efficient encoding and decoding algorithms. We set up a fully compliant link level 5G-New Radio simulator with clustered delay line channel model and we consider a realistic scenario with multiple transmitting/receiving antenna schemes and noisy downlink channel estimation. Results show that our Deep Learning approach achieves results comparable with traditional methods and can also outperform them in some conditions.","PeriodicalId":129200,"journal":{"name":"2020 IEEE Microwave Theory and Techniques in Wireless Communications (MTTW)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 IEEE Microwave Theory and Techniques in Wireless Communications (MTTW)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MTTW51045.2020.9245055","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 6
Abstract
In this paper, we study 5G Channel State Information feedback reporting. We show that a Deep Learning approach based on Convolutional Neural Networks can be used to learn efficient encoding and decoding algorithms. We set up a fully compliant link level 5G-New Radio simulator with clustered delay line channel model and we consider a realistic scenario with multiple transmitting/receiving antenna schemes and noisy downlink channel estimation. Results show that our Deep Learning approach achieves results comparable with traditional methods and can also outperform them in some conditions.