Deep learning based successive interference cancellation for the non-orthogonal downlink
Deep learning based successive interference cancellation for the non-orthogonal downlink
Non-orthogonal communications are expected to play a key role in future wireless systems. In downlink transmissions, the data symbols are broadcast from a base station to different users, which are superimposed with different power to facilitate high-integrity detection using successive interference cancellation (SIC). However, SIC requires accurate knowledge of both the channel model and channel state information (CSI), which may be difficult to acquire. We propose a deep learningaided SIC detector termed SICNet, which replaces the interference cancellation blocks of SIC by deep neural networks (DNNs). Explicitly, SICNet jointly trains its internal DNNaided blocks for inferring the soft information representing the interfering symbols in a data-driven fashion, rather than using hard-decision decoders as in classical SIC. As a result, SICNet reliably detects the superimposed symbols in the downlink of non-orthogonal systems without requiring any prior knowledge of the channel model, while being less sensitive to CSI uncertainty than its model-based counterpart. SICNet is also robust to
changes in the number of users and to their power allocation. Furthermore, SICNet learns to produce accurate soft outputs, which facilitates improved soft-input error correction decoding compared to model-based SIC. Finally, we propose an online
training method for SICNet under block fading, which exploits the channel decoding for accurately recovering online data labels for retraining, hence, allowing it to smoothly track the fading envelope without requiring dedicated pilots. Our numerical
results show that SICNet approaches the performance of classical SIC under perfect CSI, while outperforming it under realistic CSI uncertainty.
Channel models, Data models, Decoding, Interference cancellation, Receivers, Symbols, Training, Deep learning, non-orthogonal downlink, SIC
11876-11888
Luong, Thien V
a15fc6c2-8387-4e11-aae6-64553eb9770c
Shlezinger,, Nir
63f7d0ec-fa9e-4b32-8789-1c5a465cfe8f
Xu, Chao
5710a067-6320-4f5a-8689-7881f6c46252
Hoang, Minh tiep T
79ed4c0b-02ee-420a-a4cf-eeb0c2715d76
Eldar, Yonina
47dd62b0-7bce-4f28-b0d7-0003b894acf6
Hanzo, Lajos
66e7266f-3066-4fc0-8391-e000acce71a1
1 November 2022
Luong, Thien V
a15fc6c2-8387-4e11-aae6-64553eb9770c
Shlezinger,, Nir
63f7d0ec-fa9e-4b32-8789-1c5a465cfe8f
Xu, Chao
5710a067-6320-4f5a-8689-7881f6c46252
Hoang, Minh tiep T
79ed4c0b-02ee-420a-a4cf-eeb0c2715d76
Eldar, Yonina
47dd62b0-7bce-4f28-b0d7-0003b894acf6
Hanzo, Lajos
66e7266f-3066-4fc0-8391-e000acce71a1
Luong, Thien V, Shlezinger,, Nir, Xu, Chao, Hoang, Minh tiep T, Eldar, Yonina and Hanzo, Lajos
(2022)
Deep learning based successive interference cancellation for the non-orthogonal downlink.
IEEE Transactions on Vehicular Technology, 71 (11), .
(doi:10.1109/TVT.2022.3193201).
Abstract
Non-orthogonal communications are expected to play a key role in future wireless systems. In downlink transmissions, the data symbols are broadcast from a base station to different users, which are superimposed with different power to facilitate high-integrity detection using successive interference cancellation (SIC). However, SIC requires accurate knowledge of both the channel model and channel state information (CSI), which may be difficult to acquire. We propose a deep learningaided SIC detector termed SICNet, which replaces the interference cancellation blocks of SIC by deep neural networks (DNNs). Explicitly, SICNet jointly trains its internal DNNaided blocks for inferring the soft information representing the interfering symbols in a data-driven fashion, rather than using hard-decision decoders as in classical SIC. As a result, SICNet reliably detects the superimposed symbols in the downlink of non-orthogonal systems without requiring any prior knowledge of the channel model, while being less sensitive to CSI uncertainty than its model-based counterpart. SICNet is also robust to
changes in the number of users and to their power allocation. Furthermore, SICNet learns to produce accurate soft outputs, which facilitates improved soft-input error correction decoding compared to model-based SIC. Finally, we propose an online
training method for SICNet under block fading, which exploits the channel decoding for accurately recovering online data labels for retraining, hence, allowing it to smoothly track the fading envelope without requiring dedicated pilots. Our numerical
results show that SICNet approaches the performance of classical SIC under perfect CSI, while outperforming it under realistic CSI uncertainty.
Text
SICNet
- Accepted Manuscript
More information
Accepted/In Press date: 19 July 2022
e-pub ahead of print date: 22 July 2022
Published date: 1 November 2022
Additional Information:
Funding Information:
This work was supported in part by European Union’s Horizon 2020 Research and Innovation Program under Grant 646804-ERC-COG-BNYQ and in part by Israel Science Foundation underGrant 0100101. Thework of LajosHanzowas supported in part by the Engineering and Physical Sciences Research Council Projects under Grants EP/P034284/1 and EP/P003990/1 (COALESCE) and in part by the European Research Council’s Advanced Fellow Grant QuantCom under Grant 789028
Publisher Copyright:
© 1967-2012 IEEE.
Keywords:
Channel models, Data models, Decoding, Interference cancellation, Receivers, Symbols, Training, Deep learning, non-orthogonal downlink, SIC
Identifiers
Local EPrints ID: 468666
URI: http://eprints.soton.ac.uk/id/eprint/468666
ISSN: 0018-9545
PURE UUID: 49f85f0d-6ef0-49f4-96e8-a5f23739cfb9
Catalogue record
Date deposited: 19 Aug 2022 17:14
Last modified: 18 Mar 2024 03:17
Export record
Altmetrics
Contributors
Author:
Thien V Luong
Author:
Nir Shlezinger,
Author:
Chao Xu
Author:
Minh tiep T Hoang
Author:
Yonina Eldar
Author:
Lajos Hanzo
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics