A survey on essential challenges in relay-aided D2D communication for next-generation cellular networks

Abstract

Device-to-device (D2D) communication was originally presented as an efficient solution to boost conventional cellular network performance. Since the signals degrade among devices when separated by long distances, the relay-aided D2D communication appeared to enhance the coverage quality in either one-way relaying (OWR) or two-way relaying (TWR). To achieve reliable relay-aided D2D communication, the spectrum and power resources should be allocated optimally. Also, the energy harvesting (EH) technology can play an important role to improve the system energy efficiency (EE) by exploiting the radio frequency (RF) and renewable energy (RE) sources. In this paper, a comprehensive literature on the state-of-art OWR and TWR D2D communication techniques is presented. This includes the most recent power allocation (PA), resource allocation (RA), relay selection (RS), and EH techniques. The paper also shows that machine learning (ML) is the future of sixth-generation (6G) networks due to its intelligence facilities. Accordingly, we shed the light on the most important contributions of using reinforcement learning (RL) in relay-aided D2D communication. Last but not least, we highlight the research challenges and future directions for the next generation relay-aided D2D communication.