Document Type : Review Paper
Authors
1
Department of Electronics and Communications Engineering, Faculty of Electronics, Communications and Computer Engineering, Egypt-Japan University of Science and Technology, E-JUST, Alexandra, Egypt.
2
Department of Electrical Engineering, Faculty of Engineering, Assuit University, Assuit, Egypt
Abstract
The physical layer (PHY) is fundamental to wireless communication systems, enabling
robust signal transmission in complex and dynamic environments. Traditional PHY designs follow a
block-based, model-driven approach where components—such as channel coding, modulation, and
channel estimation—are optimized independently under simplified assumptions, often leading to
performance degradation in real-world scenarios. Deep Learning (DL) offers a data-driven alternative
capable of learning complex, nonlinear mappings directly from data, improving adaptability and
accuracy under diverse channel conditions. This paper reviews recent advances (2020–2025) in applying
DL to enhance key PHY functions, including channel estimation, signal detection, modulation
classification, coding/decoding, beamforming, and physical layer security. We examine various neural
architectures—such as CNNs, RNNs, autoencoders, GANs, and reinforcement learning agents—
highlighting their roles in next-generation networks (5G, B5G, 6G). While DL demonstrates superior
performance over conventional methods in adaptability, spectral efficiency, and robustness, challenges
remain in computational complexity, interpretability, training data scarcity, and generalization across
environments. The review synthesizes state-of-the-art methods, identifies open issues, and outlines
future research trends—such as model-driven DL, transfer/meta-learning, edge intelligence, and crosslayer optimization—towards building intelligent, adaptive, and scalable PHY designs for future wireless
systems.
Keywords
Main Subjects
)
View on SCiNiTO