TY - GEN
T1 - A deep reinforcement learning approach to fair distributed dynamic spectrum access
AU - Jalil, Syed Qaisar
AU - Rehmani, Mubashir Husain
AU - Chalup, Stephan
N1 - Funding Information:
SQJ was supported by a UNRSC50:50 PhD scholarship at the University of Newcastle, Australia.
Publisher Copyright:
© 2020 ACM.
PY - 2020/12/7
Y1 - 2020/12/7
N2 - This paper investigates the task how to achieve fairness in distributed dynamic spectrum access (DSA). Specifically, we consider a cognitive radio network scenario with multiple primary users (PUs) and secondary users (SUs). Each PU operates in a licensed channel. We assume that there is no coordination between PUs and SUs, and no coordination among SUs. The key challenges for SUs are to: (1) avoid collisions with PUs, (2) avoid collisions with other SUs, (3) fair access of spectrum resources in an uncoordinated system, (4) deal with different PU activity patterns, (5) deal with spectrum sensing errors. To address these challenges, we propose a deep reinforcement learning (DRL) approach and an associated reward function to achieve fair access to spectrum resources. Specifically, we use the method of Dueling Double Deep Q-Networks with Prioritised Experience Replay (D3QN-PER) as DRL algorithm for each SU. In our simulation experiments, we demonstrate that the proposed approach performs better than existing DRL methods.
AB - This paper investigates the task how to achieve fairness in distributed dynamic spectrum access (DSA). Specifically, we consider a cognitive radio network scenario with multiple primary users (PUs) and secondary users (SUs). Each PU operates in a licensed channel. We assume that there is no coordination between PUs and SUs, and no coordination among SUs. The key challenges for SUs are to: (1) avoid collisions with PUs, (2) avoid collisions with other SUs, (3) fair access of spectrum resources in an uncoordinated system, (4) deal with different PU activity patterns, (5) deal with spectrum sensing errors. To address these challenges, we propose a deep reinforcement learning (DRL) approach and an associated reward function to achieve fair access to spectrum resources. Specifically, we use the method of Dueling Double Deep Q-Networks with Prioritised Experience Replay (D3QN-PER) as DRL algorithm for each SU. In our simulation experiments, we demonstrate that the proposed approach performs better than existing DRL methods.
KW - Cognitive radio
KW - Distributed dynamic spectrum access
KW - Multi-agent deep reinforcement learning
UR - http://www.scopus.com/inward/record.url?scp=85112700009&partnerID=8YFLogxK
U2 - 10.1145/3448891.3448935
DO - 10.1145/3448891.3448935
M3 - Conference contribution
AN - SCOPUS:85112700009
T3 - ACM International Conference Proceeding Series
SP - 236
EP - 244
BT - Proceedings of the 17th EAI International Conference on Mobile and Ubiquitous Systems
PB - Association for Computing Machinery (ACM)
Y2 - 7 December 2020 through 9 December 2020
ER -