Deep learning aided LLR correction improves the performance of iterative MIMO receivers
Deep learning aided LLR correction improves the performance of iterative MIMO receivers
A Deep Learning (DL) aided Logarithmic Likelihood Ratio (LLR) correction method is proposed for improving the performance of Multiple-Input Multiple-Output (MIMO) receivers, where it is typical to adopt reduced-complexity algorithms for avoiding the excessive complexity of optimal full-search algorithms. These sub-optimal techniques typically express the probabilities of the detected bits using LLRs that often have values that are not consistent with their true reliability, either expressing too much confidence or not enough confidence in the value of the corresponding bits, leading to performance degradation. To circumvent this problem, a Deep Neural Network (DNN) is trained for detecting and correcting both over-confident and under-confident LLRs. We demonstrate that the complexity of employing the DL-aided technique is relatively low compared to the popular reduced-complexity receiver detector techniques since it only depends on a small number of real-valued inputs. Furthermore, the proposed approach is applicable to a wild variety of iterative receivers as demonstrated in the context of an iterative detection and decoding aided MIMO system, which uses a low-complexity Smart Ordering and Candidate Adding (SOCA) scheme for MIMO detection and Low-Density Parity Check (LDPC) codes for channel coding. We adopt Extrinsic Information Transfer (EXIT) charts for quantifying the Mutual Information (MI) and show that our DL method significantly improves the BLock Error Rate (BLER). Explicitly, we demonstrate that about 0.9 dB gain can be achieved at a BLER of 0.001 by employing the proposed DL-aided LLR correction method, at the modest cost of increasing the complexity by 16% compared to a benchmarker dispensing with LLR correction.
Benchmark testing, Complexity theory, DL, Detectors, Iterative decoding, Iterative methods, LDPC codes, LLR, MIMO, MIMO communication, Training, iterative detection and decoding
1 - 15
Chen, Jue
14b8e7c8-7f5e-4e68-a250-fd0989e1567b
Wang, Tsang Yi
7f1c0642-9107-4096-b255-799aff0b3176
Wu, Jwo Yuh
1c95bdaf-16e4-4c34-85b7-2df0eb2a1c0e
Li, Chih Peng
aa5cdbec-f67a-41c7-8b87-037db1ae69e3
Ng, Soon Xin
e19a63b0-0f12-4591-ab5f-554820d5f78c
Maunder, Robert G.
76099323-7d58-4732-a98f-22a662ccba6c
Hanzo, Lajos
66e7266f-3066-4fc0-8391-e000acce71a1
Chen, Jue
14b8e7c8-7f5e-4e68-a250-fd0989e1567b
Wang, Tsang Yi
7f1c0642-9107-4096-b255-799aff0b3176
Wu, Jwo Yuh
1c95bdaf-16e4-4c34-85b7-2df0eb2a1c0e
Li, Chih Peng
aa5cdbec-f67a-41c7-8b87-037db1ae69e3
Ng, Soon Xin
e19a63b0-0f12-4591-ab5f-554820d5f78c
Maunder, Robert G.
76099323-7d58-4732-a98f-22a662ccba6c
Hanzo, Lajos
66e7266f-3066-4fc0-8391-e000acce71a1
Chen, Jue, Wang, Tsang Yi, Wu, Jwo Yuh, Li, Chih Peng, Ng, Soon Xin, Maunder, Robert G. and Hanzo, Lajos
(2023)
Deep learning aided LLR correction improves the performance of iterative MIMO receivers.
IEEE Transactions on Vehicular Technology, .
(doi:10.1109/TVT.2023.3312029).
Abstract
A Deep Learning (DL) aided Logarithmic Likelihood Ratio (LLR) correction method is proposed for improving the performance of Multiple-Input Multiple-Output (MIMO) receivers, where it is typical to adopt reduced-complexity algorithms for avoiding the excessive complexity of optimal full-search algorithms. These sub-optimal techniques typically express the probabilities of the detected bits using LLRs that often have values that are not consistent with their true reliability, either expressing too much confidence or not enough confidence in the value of the corresponding bits, leading to performance degradation. To circumvent this problem, a Deep Neural Network (DNN) is trained for detecting and correcting both over-confident and under-confident LLRs. We demonstrate that the complexity of employing the DL-aided technique is relatively low compared to the popular reduced-complexity receiver detector techniques since it only depends on a small number of real-valued inputs. Furthermore, the proposed approach is applicable to a wild variety of iterative receivers as demonstrated in the context of an iterative detection and decoding aided MIMO system, which uses a low-complexity Smart Ordering and Candidate Adding (SOCA) scheme for MIMO detection and Low-Density Parity Check (LDPC) codes for channel coding. We adopt Extrinsic Information Transfer (EXIT) charts for quantifying the Mutual Information (MI) and show that our DL method significantly improves the BLock Error Rate (BLER). Explicitly, we demonstrate that about 0.9 dB gain can be achieved at a BLER of 0.001 by employing the proposed DL-aided LLR correction method, at the modest cost of increasing the complexity by 16% compared to a benchmarker dispensing with LLR correction.
Text
Deep Learning Aided LLR Correction Improves the Performance of Iterative MIMO Receivers
- Accepted Manuscript
More information
Accepted/In Press date: 31 August 2023
e-pub ahead of print date: 11 September 2023
Additional Information:
Publisher Copyright:
IEEE
Keywords:
Benchmark testing, Complexity theory, DL, Detectors, Iterative decoding, Iterative methods, LDPC codes, LLR, MIMO, MIMO communication, Training, iterative detection and decoding
Identifiers
Local EPrints ID: 481845
URI: http://eprints.soton.ac.uk/id/eprint/481845
ISSN: 0018-9545
PURE UUID: 64442ca4-1d1c-41be-a2dc-20606d6933a9
Catalogue record
Date deposited: 11 Sep 2023 17:00
Last modified: 18 Mar 2024 05:15
Export record
Altmetrics
Contributors
Author:
Jue Chen
Author:
Tsang Yi Wang
Author:
Jwo Yuh Wu
Author:
Chih Peng Li
Author:
Soon Xin Ng
Author:
Robert G. Maunder
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