Denoising and Augmentation: A Dual Use of Diffusion Model for Enhanced CSI Recovery
China Mobile Research Institute(中国移动研究院);Beijing University of Posts and Telecommunications(北京邮电大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州))
本通信快报提出了一种基于去噪扩散概率模型(DDPM,denoising diffusion probabilistic model)的信道估计算法的双重应用方案,该算法整合了数据去噪与数据增强功能。去噪环节用于解决导频位置原始信号中存在的严重噪声问题,这类噪声会降低信道估计精度;研究提出了一种无监督结构,无需纯净信道信息的先验知识即可对现场数据进行去噪处理。数据增强环节十分关键,因为用于信道状态信息(CSI,channel state information)估计的深度学习(DL,deep learning)网络训练对数据量需求较大;该网络通过调整反向步骤生成新的信道数据,丰富训练数据集。为应对通信数据中信噪比(SNR,signal-to-noise ratios)多变的问题,研究提出了一种分段前向策略以提升DDPM的收敛精度。链路级仿真结果表明,与现有基准方法相比,所提方案在精度与计算成本之间实现了更优的权衡。
英文摘要:
This letter introduces a dual application of denoising diffusion probabilistic model (DDPM)-based channel estimation algorithm integrating data denoising and augmentation. Denoising addresses the severe noise in raw signals at pilot locations, which can impair channel estimation accuracy. An unsupervised structure is proposed to clean field data without prior knowledge of pure channel information. Data augmentation is crucial due to the data-intensive nature of training deep learning (DL) networks for channel state information (CSI) estimation. The network generates new channel data by adjusting reverse steps, enriching the training dataset. To manage varying signal-to-noise ratios (SNRs) in communication data, a piecewise forward strategy is proposed to enhance the DDPM convergence precision. The link-level simulations indicate that the proposed scheme achieves a superior tradeoff between precision and computational cost compared to existing benchmarks.
ReFLEX: Length-Generalizable CSI Denoising for MIMO-OFDM via Relative-Frequency Bias
Moscow Institute of Physics and Technology (State University)(莫斯科物理技术学院(国家大学));Yangtze Delta Region Academy(长江三角洲地区研究院);Beijing Institute of Technology(北京理工大学);School of Interdisciplinary Science(交叉科学学院)
Structured Pose-Conditioned Flow Matching for Generative 5G CSI Augmentation
University of Science and Technology Beijing(北京科技大学);Sony China Research Laboratory(索尼中国研究院);The Hong Kong University of Science and Technology(香港科技大学);University of Exeter(埃克塞特大学)
Deep CSI Feedback for FDD Massive MIMO Systems: A Curvelet Learning Approach
School of Information and Communication Engineering, Xi’an Jiaotong University(西安交通大学信息与通信工程学院);School of Information Science and Technology, Jinan University(暨南大学信息科学技术学院);Laboratory of Electromagnetic Space Cognition and Intelligent Control Technology(电磁空间认知与智能控制技术实验室)
WiFo-INR: A Wireless Foundation Model Based on Implicit Neural Representations
State Key Laboratory of Photonics and Communications, School of Electronics, Peking University(光电通信国家重点实验室,电子学院,北京大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州))
APEG: Adaptive Physical Layer Authentication with Channel Extrapolation and Generative AI
State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications(北京邮电大学网络与交换技术国家重点实验室);College of Computing and Data Science, Nanyang Technological University(南洋理工大学计算与数据科学学院)
Sparse Channel Acquisition in Aging Fluid Antenna Systems: A Reliability-Calibrated Posterior Reconstruction Approach
School of Electronic and Information Engineering, South China University of Technology(华南理工大学电子与信息工程学院);School of Artificial Intelligence, Shenzhen Technology University(深圳职业技术大学人工智能学院);School of Science and Technology, Hong Kong Metropolitan University(香港都会大学理学院);Department of Electrical and Computer Engineering, The University of Texas at Dallas(德克萨斯大学达拉斯分校电气与计算机工程系)
A Novel Data Augmentation Strategy for Robust Deep Learning Classification of Biomedical Time-Series Data: Application to ECG and EEG Analysis
Computer Science Department, College of Science, University of Zakho(扎赫大学科学学院计算机科学系);Department of Computer Science and Information Technology, The American University of Kurdistan(库尔德斯坦美国大学计算机科学与信息技术系);PRIME Lab, Scientific Research Center, University of Zakho(扎赫大学科学研究中心PRIME实验室)