Rate Adaptation of D2D Underlaying Downlink Massive MIMO Networks with Reinforcement Learning

Zhang, Zezhong; Wang, Rui; Li, Yang

10.1109/GLOCOM.2018.8647257

2018

2334-0983

978-1-5386-4728-8

2018 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM)

1-7

9-13 Dec. 2018

Abu Dhabi, United arab emirates

345 E 47TH ST, NEW YORK, NY 10017 USA

IEEE

In this paper, a novel learning-based rate adaptation mechanism is proposed for downlink massive multiple-input-multiple-output (MIMO) networks with underlaid device-to-device (D2D) links, where neither the instantaneous channel state information of interfering channels nor their statistics is available at the downlink or D2D transmitters. Specifically, a coordinated scheme, where the D2D receivers join the uplink channel estimation of the associated cell, is investigated. The geographic distributions of selected downlink and D2D users are modeled as stationary and ergodic stochastic processes with unknown statistics. In order to facilitate robust rate allocation, we first derive the asymptotic signal-to-interference-and-noise ratio (SINR) for both downlink and D2D links, and show that their distributions due to unknown interfering channel can be approximated by Gaussian random variables. Then distributive reinforcement learning algorithms are proposed to evaluate the means and variances of these random variables. As a result, the base stations (BSs) and D2D transmitters could determine the transmission rates with a tolerable target packet outage probability.

Device-to-device communication Downlink Receivers Interference Transmitters Signal to noise ratio Uplink

第一 ; 通讯

英语

CPCI-S ; EI

Shenzhen Science and Technology Innovation Committee[JCYJ2016033111545794]

Engineering ; Telecommunications

Engineering, Electrical & Electronic ; Telecommunications

WOS:000465774300109

20191306709123

Channel state information ; Communication channels (information theory) ; Geographical distribution ; Learning algorithms ; Machine learning ; MIMO systems ; Random processes ; Random variables ; Signal to noise ratio ; Stochastic systems ; Transmitters ; Trellis codes

Surveying:405.3 ; Information Theory and Signal Processing:716.1 ; Data Processing and Image Processing:723.2 ; Artificial Intelligence:723.4 ; Probability Theory:922.1 ; Systems Science:961

Web of Science

Southern Univ Sci & Technol, Shenzhen, Peoples R China

Zhang, Zezhong,Wang, Rui,Li, Yang. Rate Adaptation of D2D Underlaying Downlink Massive MIMO Networks with Reinforcement Learning[C]. 345 E 47TH ST, NEW YORK, NY 10017 USA:IEEE,2018:1-7.