Multi-Head Transformer Architecture with Higher Dimensional Feature Representation for Massive MIMO CSI Feedback
Source abstract
Abstract
To achieve the anticipated performance of massive multiple input multiple output (MIMO) systems in wireless communication, it is imperative that the user equipment (UE) accurately feeds the channel state information (CSI) back to the base station (BS) along the uplink. To reduce the feedback overhead, an increasing number of deep learning (DL)-based networks have emerged, aimed at compressing and subsequently recovering CSI. Various novel structures are introduced, among which Transformer architecture has enabled a new level of precision in CSI feedback. In this paper, we propose a new method named TransNet+ built upon the Transformer-based TransNet by updating the multi-head attention layer and implementing an improved training scheme. The simulation results demonstrate that TransNet+ outperforms existing methods in terms of recovery accuracy and achieves state-of-the-art.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Antenna Design and Optimization · Advanced Adaptive Filtering Techniques · Full-Duplex Wireless Communications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Electronics and Information Engineering, Tongji University, Shanghai 201800, China
Funding
Future Network Innovation Research and Application Project (2021)
References (42)
Show source-provided references
1. Han, F., Zeng, J., Zheng, L., Zhang, H., and Wang, J. (2024). Sensing and Deep CNN-Assisted Semi-Blind Detection for Multi-User Massive MIMO Communications. Remote Sens., 16.10.3390/rs16020247 ↗
2. Lin, W.-Y., Chang, T.-H., and Tseng, S.-M. (2023). Deep Learning-Based Cross-Layer Power Allocation for Downlink Cell-Free Massive Multiple-Input–Multiple-Output Video Communication Systems. Symmetry, 15.10.3390/sym15111968 ↗
3. Pan, F., Zhao, X., Zhang, B., Xiang, P., Hu, M., and Gao, X. (2023). CSI Feedback Model Based on Multi-Source Characterization in FDD Systems. Sensors, 23.10.3390/s23198139 ↗
4. Liu, Q., Sun, J., and Wang, P. (2023). Uplink Assisted MIMO Channel Feedback Method Based on Deep Learning. Entropy, 25.10.3390/e25081131 ↗
5. Riviello, D.G., Tuninato, R., Zimaglia, E., Fantini, R., and Garello, R. (2023). Implementation of Deep-Learning-Based CSI Feedback Reporting on 5G NR-Compliant Link-Level Simulator. Sensors, 23.10.3390/s23020910 ↗
6. Sun, Q., Zhao, H., Wang, J., and Chen, W. (2022). Deep Learning-Based Joint CSI Feedback and Hybrid Precoding in FDD mmWave Massive MIMO Systems. Entropy, 24.10.3390/e24040441 ↗
7. Naser, M.A., Abdul-Hadi, A.M., Alsabah, M., Mahmmod, B.M., Majeed, A., and Abdulhussain, S.H. (2023). Downlink Training Sequence Design Based on Waterfilling Solution for Low-Latency FDD Massive MIMO Communications Systems. Electronics, 12.10.3390/electronics12112494 ↗
8. A novel CSI feedback approach for massive MIMO using LSTM-attention CNN10.1109/ACCESS.2020.2963896 ↗
9. Manasa, B.M.R., Pakala, V., Chinthaginjala, R., Ayadi, M., Hamdi, M., and Ksibi, A. (2023). A Novel Channel Estimation Framework in MIMO Using Serial Cascaded Multiscale Autoencoder and Attention LSTM with Hybrid Heuristic Algorithm. Sensors, 23.10.3390/s23229154 ↗
10. A novel approach using convolutional transformer for massive MIMO CSI feedback10.1109/LWC.2022.3153085 ↗
11. TransNet: Full attention network for CSI feedback in FDD massive MIMO system10.1109/LWC.2022.3149416 ↗
12. Lu, Z., Wang, J., and Song, J. (2020, January 7–11). Multi-resolution CSI feedback with deep learning in massive MIMO system. Proceedings of the 2020 IEEE International Conference on Communications (ICC), Dublin, Ireland.10.1109/ICC40277.2020.9149229 ↗
13. Chen, J., and Mei, M. (2023). Numerical Analysis of Low-Cost Recognition of Tunnel Cracks with Compressive Sensing along the Railway. Appl. Sci., 13.10.3390/app132413007 ↗
14. Deep learning for massive MIMO CSI feedback10.1109/LWC.2018.2818160 ↗
15. Sharma, S., and Yoon, W. (2023). Energy Efficient Power Allocation in Massive MIMO Based on Parameterized Deep DQN. Electronics, 12.10.20944/preprints202310.0066.v1 ↗
16. Zhang, Y., and Luo, Z. (2023). A Review of Research on Spectrum Sensing Based on Deep Learning. Electronics, 12.10.3390/electronics12214514 ↗
17. Bit-level optimized neural network for multi-antenna channel quantization10.1109/LWC.2019.2942908 ↗
18. CLNet: Complex input lightweight neural network designed for massive MIMO CSI feedback10.1109/LWC.2021.3100493 ↗
19. Binary neural network aided CSI feedback in massive MIMO system10.1109/LWC.2021.3064963 ↗
20. Convolutional neural network based multiple-rate compressive sensing for massive MIMO CSI feedback: Design, simulation, and analysis10.1109/TWC.2020.2968430 ↗
21. Deep learning-based CSI feedback approach for time-varying massive MIMO channels10.1109/LWC.2018.2874264 ↗
22. Cai, Q., Dong, C., and Niu, K. (2019, January 15–18). Attention model for massive MIMO CSI compression feedback and recovery. Proceedings of the 2019 IEEE Wireless Communications and Networking Conference (WCNC), Marrakesh, Morocco.10.1109/WCNC.2019.8885897 ↗
23. SALDR: Joint self-attention learning and dense refine for massive MIMO CSI feedback with multiple compression ratio10.1109/LWC.2021.3085317 ↗
24. Machine learning-based adaptive CSI feedback interval10.1016/j.icte.2021.11.007 ↗
25. Exploiting bi-directional channel reciprocity in deep learning for low rate massive MIMO CSI feedback10.1109/LWC.2019.2898662 ↗
26. Compressive sampled CSI feedback method based on deep learning for FDD massive MIMO systems10.1109/TCOMM.2021.3086525 ↗
27. Distributed deep convolutional compression for massive MIMO CSI feedback10.1109/TWC.2020.3043502 ↗
28. A Markovian model-driven deep learning framework for massive MIMO CSI feedback10.1109/TWC.2021.3103120 ↗
29. CAnet: Uplink-aided downlink channel acquisition in FDD massive MIMO using deep learning10.1109/TCOMM.2021.3120294 ↗
30. Deep learning-based implicit CSI feedback in massive MIMO10.1109/TCOMM.2021.3138097 ↗
31. Deep learning for CSI feedback based on superimposed coding10.1109/ACCESS.2019.2928049 ↗
32. Feedback of downlink channel state information based on superimposed coding10.1109/LCOMM.2007.061534 ↗
33. Guo, J., Chen, T., and Jin, S. (2023). Deep learning for joint channel estimation and feedback in massive MIMO systems. Digit. Commun. Netw.10.1016/j.dcan.2023.01.011 ↗
34. Deep learning-based limited feedback designs for MIMO systems10.1109/LWC.2019.2962114 ↗
35. Deep learning-based denoise network for CSI feedback in FDD massive MIMO systems10.1109/LCOMM.2020.2989499 ↗
36. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., and Gomez, A.N. (2017, January 4–9). Attention is all you need. Proceedings of the 31st International Conference on Neural Information Processing Systems (ICONIP), Long Beach, CA, USA.
37. Xue, J., Chen, X., Chi, Q., and Xiao, W. (2024). Online Learning-Based Adaptive Device-Free Localization in Time-Varying Indoor Environment. Appl. Sci, 14.10.3390/app14020643 ↗
38. The cost 2100 MIMO channel model10.1109/MWC.2012.6393523 ↗
39. Overview of deep learning-based CSI feedback in massive MIMO systems10.1109/TCOMM.2022.3217777 ↗
40. MRFNet: A deep learning-based CSI feedback approach of massive MIMO systems10.1109/LCOMM.2021.3099841 ↗
41. A lightweight deep network for efficient CSI feedback in massive MIMO systems10.1109/LWC.2021.3083331 ↗
42. Deep learning based CSI compression and quantization with high compression ratios in FDD massive MIMO systems10.1109/LWC.2021.3092947 ↗