King's College London

Research portal

Learning Based Signal Detection for MIMO Systems with Unknown Noise Statistics

Research output: Contribution to journalArticlepeer-review

Ke He, Le He, Lisheng Fan, Yansha Deng, George K. Karagiannidis, Arumugam Nallanathan

Original languageEnglish
Article number9353547
Pages (from-to)3025-3038
Number of pages14
Issue number5
Accepted/In press2021
PublishedMay 2021

Bibliographical note

Funding Information: Manuscript received August 11, 2020; revised December 27, 2020; accepted February 4, 2021. Date of publication February 12, 2021; date of current version May 18, 2021. This work was supported by the NSFC (Nos. 61871139/61801132), by the International Science and Technology Cooperation Projects of Guangdong Province (No. 2020A0505100060), by Natural Science Foundation of Guangdong Province (Nos. 2017A030308006/2018A030310338/2020A1515010484), by the Science and Technology Program of Guangzhou (No. 201807010103), and by the research program of Guangzhou University (No. YK2020008). The work of George K. Karagiannidis has been co-financed by the European Union and Greek national funds through the Competitiveness, Entrepreneurship and Innovation Operational Program (EPAnEK), under the special actions AQUACULTURE–INDUSTRIAL MATERIALS–OPEN INNOVATION IN CULTURE (project code: T6YBP-00134). The associate editor coordinating the review of this article and approving it for publication was C.-H. Lee. (Corresponding author: Lisheng Fan.) Ke He, Le He, and Lisheng Fan are with the School of Computer Science and Cyber Engineering, Guangzhou University, Guangzhou 510006, China (e-mail:;; lsfan@ Publisher Copyright: © 1972-2012 IEEE. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.

King's Authors


This paper aims to devise a generalized maximum likelihood (ML) estimator to robustly detect signals with unknown noise statistics in multiple-input multiple-output (MIMO) systems. In practice, there is little or even no statistical knowledge on the system noise, which in many cases is non-Gaussian, impulsive and not analyzable. Existing detection methods have mainly focused on specific noise models, which are not robust enough with unknown noise statistics. To tackle this issue, we propose a novel ML detection framework to effectively recover the desired signal. Our framework is a fully probabilistic one that can efficiently approximate the unknown noise distribution through a normalizing flow. Importantly, this framework is driven by an unsupervised learning approach, where only the noise samples are required. To reduce the computational complexity, we further present a low-complexity version of the framework, by utilizing an initial estimation to reduce the search space. Simulation results show that our framework outperforms other existing algorithms in terms of bit error rate (BER) in non-analytical noise environments, while it can reach the ML performance bound in analytical noise environments.

View graph of relations

© 2020 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454