Enhanced deep soft interference cancellation for multiuser symbol detection

다국어 초록 (Multilingual Abstract)

The detection of all the symbols transmitted simultaneously in multiuser systems using limited wireless resources is challenging. Traditional model-based methods show high performance with perfect channel state information (CSI); however, severe performance degradation will occur if perfect CSI cannot be acquired. In contrast, data-driven methods perform slightly worse than model-based methods in terms of symbol error ratio performance in perfect CSI states; however, they are also able to overcome extreme performance degradation in imperfect CSI states. This study proposes a novel deep learning-based method by improving a state-of-the-art data-driven technique called deep soft interference cancellation (DSIC). The enhanced DSIC (EDSIC) method detects multiuser symbols in a fully sequential manner and ses an efficient neural network structure to ensure high performance. Additionally, error-propagation mitigation techniques are used to ensure robustness against channel uncertainty. The EDSIC guarantees a performance that is very close to the optimal performance of the existing model-based methods in perfect CSI environments and the best performance in imperfect CSI environments.

참고문헌 (Reference)

1 F. Liang, "Towards optimal power control via ensembling deep neural networks" 68 (68): 1760-1776, 2020

2 Lee, Howon ; Lee, Byungju ; Yang, Heecheol ; Kim, Junghyun ; Kim, Seungnyun ; Shin, Wonjae ; Shim, Byonghyo ; Poor, H. Vincent, "Towards 6G Hyper-Connectivity: Vision, Challenges, and Key Enabling Technologies" 한국통신학회 25 (25): 344-354, 2023

3 H. Kim, "Physical layer communication via deep learning" 1 (1): 5-18, 2020

4 X. Chen, "On the design of massive nonorthogonal multiple access with imperfect successive interference cancellation" 67 (67): 2539-2551, 2019

5 D. Silver, "Mastering the game of Go without human knowledge" 550 (550): 354-359, 2017

6 D. Gündüz, "Machine learning in the air" 37 (37): 2184-2199, 2019

7 H. Sun, "Learning to optimize : training deep neural networks for interference management" 66 (66): 5438-5453, 2018

8 S. Zhang, "Learn codes : inventing lowlatency codes via recurrent neural networks" 1 (1): 207-216, 2020

9 W. -J. Choi, "Iterative soft interference cancellation for multiple antenna systems" 304-309, 2000

10 N. Shlezinger, "Hybrid beamforming based on an unsupervised deep learning network for downlink channels with imperfect CSI" 11 (11): 1543-1547, 2022

11 Y. Shen, "Graph neural networks for scalable radio resource management : architecture design and theoretical analysis" 39 (39): 101-115, 2021

12 O. Vinyals, "Grandmaster level in Starcraft II using multi-agent reinforcement learning" 575 (575): 350-354, 2019

13 N. Shlezinger, "DeepSIC : deep soft interference cancellation for multiuser MIMO detection" 20 (20): 1349-1362, 2021

14 M. Honkala, "DeepRx : fully convolutional deep learning receiver" 20 (20): 3925-3940, 2021

15 W. Lee, "Deep-learning-assisted wireless-powered secure communications with imperfect channel state information" 9 (9): 11464-11476, 2022

16 W. Lee, "Deep power control : transmit power control scheme based on convolutional neural network" 22 (22): 1276-1279, 2018

17 J. Kim, "Deep learningassisted multi-dimensional modulation and resource mapping for advanced OFDM systems" 1-6, 2018

18 H. Ye, "Deep learning-based end-to-end wireless communication systems with conditional gans as unknown channels" 19 (19): 3133-3143, 2020

19 E. Nachmani, "Deep learning methods for improved decoding of linear codes" 12 (12): 119-131, 2018

20 Q. Mao, "Deep learning for intelligent wireless networks : a comprehensive survey" 20 (20): 2595-2621, 2018

21 T. V. Luong, "Deep learning based successive interference cancellation for the non-orthogonal downlink" 71 (71): 11876-11888, 2022

22 S. Dörner, "Deep learning based communication over the air" 12 (12): 132-143, 2018

23 Y. LeCun, "Deep learning" 521 (521): 436-444, 2015

24 I. Goodfellow, "Deep learning" MIT Press 2016

25 T. O’Shea, "An introduction to deep learning for the physical layer" 3 (3): 563-575, 2017

26 S. Han, "A new design of channel denoiser using residual autoencoder" 59 (59): 1-3, 2023

27 C. Lin, "A deep learning approach for MIMO-NOMA downlink signal detection" 19 (19): 1-22, 2021

28 "3GPP, Technical Specification Group Radio Access Network;NR; multiplexing and channel coding; (Release 17), Technical Specification (TS) 38.212, 3rd Generation Partnership Project (3GPP), version 17.4.0"