Transfer Generative Adversarial Networks (T-GAN)-based Terahertz Temporal Channel Modeling and Generating

被引:2
作者
Hu, Zhengdong [1 ]
Li, Yuanbo [1 ]
Han, Chong [1 ]
机构
[1] Shanghai Jiao Tong Univ, Shanghai, Peoples R China
来源
PROCEEDINGS OF THE 2023 THE 7TH ACM WORKSHOP ON MILLIMETER-WAVE AND TERAHERTZ NETWORKS AND SENSING SYSTEMS, MMNETS 2023 | 2023年
关键词
Terahertz; channel modeling; transfer learning; generative adversarial networks;
D O I
10.1145/3615360.3625091
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Terahertz (THz) communications are a promising technology for 6G and beyond wireless systems, offering ultra-broad bandwidth and thus data rates of Terabit-per-second (Tbps). However, accurate channel modeling and characterization are fundamental for the design of THz communications. Relying on channel measurements, traditional statistical channel modeling methods suffer from low accuracy due to the assumed certain distributions and empirical parameters. Moreover, acquiring extensive channel measurement is time-consuming and expensive in the THz band. To address these challenges, a transfer generative adversarial network (T-GAN) based modeling method is proposed in the THz band, which exploits the advantage of GAN in modeling the complex distribution. Moreover, the transfer learning technique is introduced in T-GAN, which transfers the knowledge stored in a pre-trained model based on simulated data, to a new model based on a small amount of measured data. The simulation data is generated by the standard channel model from 3rd generation partnerships project (3GPP), which contains the knowledge that can be transferred to reduce the demand of measurement data and improve the accuracy of T-GAN. Experimental results reveal that the distribution of power delay profiles (PDPs) generated by the proposed T-GAN method shows good agreement with measurement. Moreover, T-GAN achieves good performance in channel modeling, with 9 dB improved root-mean-square error (RMSE) and higher Structure Similarity Index Measure (SSIM), compared with traditional 3GPP method.
引用
收藏
页码:7 / 12
页数:6
相关论文
共 50 条
[21]   CCDWT-GAN: Generative Adversarial Networks Based on Color Channel Using Discrete Wavelet Transform for Document Image Binarization [J].
Ju, Rui-Yang ;
Lin, Yu-Shian ;
Chiang, Jen-Shiun ;
Chen, Chih-Chia ;
Chen, Wei-Han ;
Chien, Chun-Tse .
PRICAI 2023: TRENDS IN ARTIFICIAL INTELLIGENCE, PT I, 2024, 14325 :186-198
[22]   Single-Channel Speech Quality Enhancement in Mobile Networks Based on Generative Adversarial Networks [J].
Wu, Guifen ;
Herencsar, Norbert .
MOBILE NETWORKS & APPLICATIONS, 2024, 29 (06) :1909-1923
[23]   Anime Image Style Transfer Algorithm Based on Improved Generative Adversarial Networks [J].
Li, Yunhong ;
Zhu, Jingkun ;
Liu, Xingrui ;
Chen, Jinni ;
Su, Xueping .
Beijing Youdian Daxue Xuebao/Journal of Beijing University of Posts and Telecommunications, 2024, 47 (04) :117-123
[24]   Curb-GAN: Conditional Urban Traffic Estimation through Spatio-Temporal Generative Adversarial Networks [J].
Zhang, Yingxue ;
Li, Yanhua ;
Zhou, Xun ;
Kong, Xiangnan ;
Luo, Jun .
KDD '20: PROCEEDINGS OF THE 26TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY & DATA MINING, 2020, :842-852
[25]   Geological Facies modeling based on progressive growing of generative adversarial networks (GANs) [J].
Suihong Song ;
Tapan Mukerji ;
Jiagen Hou .
Computational Geosciences, 2021, 25 :1251-1273
[26]   Geological Facies modeling based on progressive growing of generative adversarial networks (GANs) [J].
Song, Suihong ;
Mukerji, Tapan ;
Hou, Jiagen .
COMPUTATIONAL GEOSCIENCES, 2021, 25 (03) :1251-1273
[27]   A survey on modeling for behaviors of complex intelligent systems based on generative adversarial networks [J].
Lv, Yali ;
Duan, Jingpu ;
Li, Xiong .
COMPUTER SCIENCE REVIEW, 2024, 52
[28]   A Generative Adversarial Networks (GAN) based efficient sampling method for inverse design of metallic glasses [J].
Xu, Xiang ;
Hu, Jingyi .
JOURNAL OF NON-CRYSTALLINE SOLIDS, 2023, 613
[29]   REG-GAN: SEMI-SUPERVISED LEARNING BASED ON GENERATIVE ADVERSARIAL NETWORKS FOR REGRESSION [J].
Rezagholizadeh, Mehdi ;
Haidar, Md Akmal .
2018 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2018, :2806-2810
[30]   IoT-GAN: Anomaly Detection for Time Series in IoT Based on Generative Adversarial Networks [J].
Chen, Xiaofei ;
Zhang, Shuo ;
Jiang, Qiao ;
Chen, Jiayuan ;
Huang, Hejiao ;
Gu, Chonglin .
ALGORITHMS AND ARCHITECTURES FOR PARALLEL PROCESSING, ICA3PP 2021, PT II, 2022, 13156 :682-694