TY - JOUR
T1 - Dynamic Spectrum Access Using Stochastic Multi-User Bandits
AU - Bande, Meghana
AU - Magesh, Akshayaa
AU - Veeravalli, Venugopal V.
N1 - Manuscript received December 7, 2020; accepted January 7, 2021. Date of publication January 13, 2021; date of current version May 10, 2021. This work was supported by the U.S. NSF SpecEES through the University of Illinois at Urbana–Champaign under Grant 1730882. The associate editor coordinating the review of this article and approving it for publication was R. Wang. (Corresponding author: Venugopal V. Veeravalli.) Meghana Bande was with the Department of Electrical and Computer Engineering, University of Illinois at Urbana–Champaign, IL 61820 USA. She is now with Qualcomm Technologies Inc., Bridgewater, NJ 08807 USA (e-mail: [email protected]).
PY - 2021/5
Y1 - 2021/5
N2 - A stochastic multi-user multi-armed bandit framework is used to develop algorithms for uncoordinated spectrum access. In contrast to prior work, it is assumed that rewards can be non-zero even under collisions, thus allowing for the number of users to be greater than the number of channels. The proposed algorithm consists of an estimation phase and an allocation phase. It is shown that if every user adopts the algorithm, the system wide regret is order-optimal of order O (log T) over a time-horizon of duration T. The regret guarantees hold for both the cases where the number of users is greater than or less than the number of channels. The algorithm is extended to the dynamic case where the number of users in the system evolves over time, and is shown to lead to sub-linear regret.
AB - A stochastic multi-user multi-armed bandit framework is used to develop algorithms for uncoordinated spectrum access. In contrast to prior work, it is assumed that rewards can be non-zero even under collisions, thus allowing for the number of users to be greater than the number of channels. The proposed algorithm consists of an estimation phase and an allocation phase. It is shown that if every user adopts the algorithm, the system wide regret is order-optimal of order O (log T) over a time-horizon of duration T. The regret guarantees hold for both the cases where the number of users is greater than or less than the number of channels. The algorithm is extended to the dynamic case where the number of users in the system evolves over time, and is shown to lead to sub-linear regret.
KW - Cognitive radio
KW - multi-armed bandits
KW - sub-linear regret
KW - uncoordinated spectrum access
UR - https://www.scopus.com/pages/publications/85099546475
UR - https://www.scopus.com/pages/publications/85099546475#tab=citedBy
U2 - 10.1109/LWC.2021.3051328
DO - 10.1109/LWC.2021.3051328
M3 - Article
AN - SCOPUS:85099546475
SN - 2162-2337
VL - 10
SP - 953
EP - 956
JO - IEEE Wireless Communications Letters
JF - IEEE Wireless Communications Letters
IS - 5
M1 - 9321391
ER -