DQ-MAC: an adaptive mac protocol switching scheme based on deep Q-learning

Zhiao Zheng; Bo Jiang; Hong XU; Chang Shi

doi:10.1117/12.2652778

6 September 2022 DQ-MAC: an adaptive mac protocol switching scheme based on deep Q-learning

Zhiao Zheng, Bo Jiang, Hong XU, Chang Shi

Proceedings Volume 12332, International Conference on Intelligent Systems, Communications, and Computer Networks (ISCCN 2022); 1233204 (2022) https://doi.org/10.1117/12.2652778
Event: International Conference on Intelligent Systems, Communications, and Computer Networks (ISCCN 2022), 2022, Chengdu, China

Abstract

MAC protocols are important to ensure the quality of service of UAV ad-hoc networks, which are becoming more and more popular. Facing the dynamically changing network transmission environment, the adaptive MAC protocol becomes an effective solution. In this paper, a MAC protocol based on deep Q-learning is designed. The system integrates a contention-based MAC protocol and a scheduling-based MAC protocol. Using the deep Q-learning approach, it switches between CSMA and TDMA according to the current state of the UAV (e.g. throughput rate, latency, etc.). Two different network scenarios were designed and simulated to evaluate the performance of the designed MAC protocols. The results show that the performance of the designed adaptive MAC protocol outperforms that of a single protocol in terms of performance such as latency and throughput.

Citation Download Citation

Zhiao Zheng, Bo Jiang, Hong XU, and Chang Shi "DQ-MAC: an adaptive mac protocol switching scheme based on deep Q-learning", Proc. SPIE 12332, International Conference on Intelligent Systems, Communications, and Computer Networks (ISCCN 2022), 1233204 (6 September 2022); https://doi.org/10.1117/12.2652778

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available