Learning-Driven Wireless Communications, towards 6G

The fifth generation (5G) of wireless communication is in its infancy, and its evolving versions will be launched over the coming years. However, according to exposing the inherent constraints of 5G and the emerging applications and services with stringent requirements e.g. latency, energy/bit, traffic capacity, peak data rate, and reliability, telecom researchers are turning their attention to conceptualize the next generation of wireless communications, i.e. 6G. In this paper, we investigate 6G challenges, requirements, and trends. Furthermore, we discuss how artificial intelligence (AI) techniques can contribute to 6G. Based on the requirements and solutions, we identify some new fascinating services and use-cases of 6G, which can not be supported by 5G appropriately. Moreover, we explain some research directions that lead to the successful conceptualization and implementation of 6G.

[1]  Matti Latva-aho,et al.  6Genesis Flagship Program: Building the Bridges Towards 6G-Enabled Wireless Smart Society and Ecosystem , 2018, 2018 IEEE 10th Latin-American Conference on Communications (LATINCOM).

[2]  Shree Krishna Sharma,et al.  Quantum Machine Learning for 6G Communication Networks: State-of-the-Art and Vision for the Future , 2019, IEEE Access.

[3]  Jakob Hoydis,et al.  An Introduction to Deep Learning for the Physical Layer , 2017, IEEE Transactions on Cognitive Communications and Networking.

[4]  Geoffrey Ye Li,et al.  Power of Deep Learning for Channel Estimation and Signal Detection in OFDM Systems , 2017, IEEE Wireless Communications Letters.

[5]  Xianbin Wang,et al.  Collaborative Distributed Q-Learning for RACH Congestion Minimization in Cellular IoT Networks , 2019, IEEE Communications Letters.

[6]  Geoffrey Ye Li,et al.  Deep Reinforcement Learning for Resource Allocation in V2V Communications , 2017, 2018 IEEE International Conference on Communications (ICC).