Deep Learning for Physical-Layer 5G Wireless Techniques: Opportunities, Challenges and Solutions

TLDR

High‑reliability and ultra‑high capacity demands drive 5G research, but conventional theories limit performance; deep learning is emerging as a promising tool to optimize wireless communications. The article reviews deep‑learning solutions for 5G and proposes efficient schemes for deep‑learning‑based 5G scenarios. The authors present key ideas for deep‑learning‑based communication methods, investigate novel NOMA, massive MIMO, and mmWave frameworks, and highlight research opportunities and challenges. The work suggests that deep‑learning‑based wireless physical‑layer frameworks will pioneer new directions in communication theory.

Abstract

The new demands for high-reliability and ultra-high capacity wireless communication have led to extensive research into 5G communications. However, current communication systems, which were designed on the basis of conventional communication theories, significantly restrict further performance improvements and lead to severe limitations. Recently, the emerging deep learning techniques have been recognized as a promising tool for handling the complicated communication systems, and their potential for optimizing wireless communications has been demonstrated. In this article, we first review the development of deep learning solutions for 5G communication, and then propose efficient schemes for deep learning-based 5G scenarios. Specifically, the key ideas for several important deep learning-based communication methods are presented along with the research opportunities and challenges. In particular, novel communication frameworks of NOMA, massive multiple-input multiple-output (MIMO), and millimeter wave (mmWave) are investigated, and their superior performances are demonstrated. We envision that the appealing deep learning- based wireless physical layer frameworks will bring a new direction in communication theories and that this work will move us forward along this road.