TY - GEN
T1 - A Context-based Long-Term Deployment Algorithm for Multiple Relaying Unmanned Aerial Vehicles
AU - Zhao, Le
AU - Yin, Yuhang
AU - Yan, Lei
AU - Lu, Jihua
AU - Lihui-Feng,
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - A paradigm for the long-term deployment of multiple relaying unmanned aerial vehicles (UAVs) is proposed. The deployment is formulated as a sequential decision-making problem under the multi-armed bandit (MAB) framework, in which the UAVs are considered as players, and the relay location of UAVs are considered as arms to be selected. Considering the limited power and recharging requirement in long-term deployment, a contextual adaptive and recharge-oriented upper confidence bound (CAR-UCB) algorithm is designed. We take the advantages of context feature, feature weight matrix and reward quantification to learn from environment and interaction history as linear parameters. The experimental results reveal that the proposed approach achieves satisfactory performance in network coverage, power efficiency and average reward in different scenarios with a sightly increased complexity.
AB - A paradigm for the long-term deployment of multiple relaying unmanned aerial vehicles (UAVs) is proposed. The deployment is formulated as a sequential decision-making problem under the multi-armed bandit (MAB) framework, in which the UAVs are considered as players, and the relay location of UAVs are considered as arms to be selected. Considering the limited power and recharging requirement in long-term deployment, a contextual adaptive and recharge-oriented upper confidence bound (CAR-UCB) algorithm is designed. We take the advantages of context feature, feature weight matrix and reward quantification to learn from environment and interaction history as linear parameters. The experimental results reveal that the proposed approach achieves satisfactory performance in network coverage, power efficiency and average reward in different scenarios with a sightly increased complexity.
KW - Unmanned aerial vehicle
KW - contextual multiarmed bandit
KW - feature weight
KW - recharge orientation
KW - reward quantification
UR - http://www.scopus.com/inward/record.url?scp=85139487327&partnerID=8YFLogxK
U2 - 10.1109/ICCC55456.2022.9880788
DO - 10.1109/ICCC55456.2022.9880788
M3 - Conference contribution
AN - SCOPUS:85139487327
T3 - 2022 IEEE/CIC International Conference on Communications in China, ICCC 2022
SP - 139
EP - 143
BT - 2022 IEEE/CIC International Conference on Communications in China, ICCC 2022
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2022 IEEE/CIC International Conference on Communications in China, ICCC 2022
Y2 - 11 August 2022 through 13 August 2022
ER -