TY - GEN
T1 - Realtime Scheduling and Power Allocation Using Deep Neural Networks
AU - Xu, Shenghe
AU - Liu, Pei
AU - Wang, Ran
AU - Panwar, Shivendra S.
N1 - Funding Information:
This work was supported by the National Science Foundation under Grant No. 1527750, as well as the NYSTAR Center for Advanced Technology in Telecommunications (CATT), and NYU WIRELESS.
Funding Information:
VII. ACKNOWLEDGEMENTS This work was supported by the National Science Foundation under Grant No. 1527750, as well as the NYSTAR Center for Advanced Technology in Telecommunications (CATT), and NYU WIRELESS.
Publisher Copyright:
© 2019 IEEE.
PY - 2019/4
Y1 - 2019/4
N2 - With the increasing number of base stations (BSs) and network densification in 5G, interference management using link scheduling and power control are vital for better utilization of radio resources. However, the complexity of solving link scheduling and the power control problem grows exponentially with the number of BS. Due to high computation time, previous methods are useful for research purposes but impractical for real time usage. In this paper we propose to use deep neural networks (DNNs) to approximate optimal link scheduling and power control for the case with multiple small cells. A deep Q-network (DQN) estimates a suitable schedule, then a DNN allocates power for the corresponding schedule. Simulation results show that compared with Geometric Programming based power allocation and exhaustive search based scheduling, the proposed method achieves over five orders of magnitude speed-up with less than nine percent performance loss, making real time usage practical.
AB - With the increasing number of base stations (BSs) and network densification in 5G, interference management using link scheduling and power control are vital for better utilization of radio resources. However, the complexity of solving link scheduling and the power control problem grows exponentially with the number of BS. Due to high computation time, previous methods are useful for research purposes but impractical for real time usage. In this paper we propose to use deep neural networks (DNNs) to approximate optimal link scheduling and power control for the case with multiple small cells. A deep Q-network (DQN) estimates a suitable schedule, then a DNN allocates power for the corresponding schedule. Simulation results show that compared with Geometric Programming based power allocation and exhaustive search based scheduling, the proposed method achieves over five orders of magnitude speed-up with less than nine percent performance loss, making real time usage practical.
KW - deep neural networks
KW - deep reinforcement learning
KW - power allocation
KW - scheduling
UR - http://www.scopus.com/inward/record.url?scp=85074787544&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85074787544&partnerID=8YFLogxK
U2 - 10.1109/WCNC.2019.8886140
DO - 10.1109/WCNC.2019.8886140
M3 - Conference contribution
AN - SCOPUS:85074787544
T3 - IEEE Wireless Communications and Networking Conference, WCNC
BT - 2019 IEEE Wireless Communications and Networking Conference, WCNC 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2019 IEEE Wireless Communications and Networking Conference, WCNC 2019
Y2 - 15 April 2019 through 19 April 2019
ER -