Non-Uniform Time-Step Deep Q-Network for Carrier-Sense Multiple Access in Heterogeneous Wireless Networks
Publication in refereed journal
Officially Accepted for Publication

Altmetrics Information

Other information
AbstractThis paper investigates CSMA protocols that employ deep reinforcement learning (DRL) techniques, referred to as carrier-sense deep-reinforcement learning multiple access (CS-DLMA). The goal of CS-DLMA is to enable efficient spectrum sharing among a group of co-located heterogeneous wireless networks. Existing CSMA protocols, such as the MAC of WiFi, are designed for a homogeneous network. Such protocols suffer from severe performance degradation in heterogeneous environments. CS-DLMA aims to circumvent this problem by making use of DRL. In particular, this paper adopts alpha-fairness as the objective of CS-DLMA. With alpha-fairness, CS-DLMA can achieve different objectives when coexisting with other MACs. A salient feature of CS-DLMA is that it can achieve these objectives without knowing the coexisting MACs. The underpinning DRL technique in CS-DLMA is deep Q-network (DQN). However, the conventional DQN algorithms are not suitable for CS-DLMA due to uniform time-step assumption. In CSMA protocols, time steps are non-uniform in that the time durations for carrier sensing and data transmission are different. This paper introduces non-uniform time-step DQN to address this issue. Our simulation results show that CS-DLMA can achieve alpha-fairness objective when coexisting with TDMA, ALOHA, and WiFi protocols. We also find that CS-DLMA is more Pareto efficient than other CSMA protocols when coexisting with WiFi.
Acceptance Date27/04/2020
All Author(s) ListY. Yu, S. C. Liew, T. Wang
Journal nameIEEE Transactions on Mobile Computing
Place of PublicationUSA
LanguagesEnglish-United States

Last updated on 2020-03-12 at 23:08