A deep learning‐based HARQ chase combining for FBMC‐OQAM systems. (26th August 2021)

• Record Type: Journal Article
• Authors: Bedoui, Abla; Et‐tolba, Mohamed
• Is Part Of: International journal of communication systems. Volume 34:Number 16(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-08-26
• Subjects: BLER -- deep learning -- FBMC‐OQAM -- fuzzy inference -- HARQ -- RMSE; Telecommunication systems -- Periodicals; 621.382
• DOI: 10.1002/dac.4957
• Languages: English
• ISSNs: 1074-5351

Abstract

Summary: Hybrid automatic repeat request (HARQ) is a promising link adaptation technique to ensure reliable data transmission in future communication systems. However, it can severely suffer from high distortions, nonlinearities, and interferences caused by the propagation environment. In this paper, we propose an efficient HARQ chase combining (HARQ‐CC) for filter bank multicarrier with offset quadrature amplitude modulation (FBMC‐OQAM) systems using deep learning to combat channel impairments. We model the HARQ‐CC system as a fuzzy inference system that deploys a deep neural network tool to detect the transmitted data packets. Simulation results prove that the proposed approach outperforms the conventional HARQ‐CC method with regard to block error rate (BLER) and root mean square error (RMSE). Abstract : Hybrid automatic repeat request (HARQ) suffers from high distortions, nonlinearities, and interferences caused by the transmission environment. To fight the premodel, we use the HARQ‐chase combining (CC) filter bank multicarrier with offset quadrature amplitude modulation (FBMC‐OQAM)‐based system as a fuzzy inference system that uses a deep neural network to efficiently detect the transmitted data packets. The proposed approach provides a considerable performance gain as compared with the conventional methods.