TY - GEN
T1 - LTE handover parameters optimization using Q-learning technique
AU - Abdelmohsen, Assem
AU - Abdelwahab, Mohamed
AU - Adel, Mohamed
AU - Saeed Darweesh, M.
AU - Mostafa, Hassan
N1 - KAUST Repository Item: Exported on 2022-06-30
Acknowledgements: This research was partially funded by ONE Lab at Cairo University, Zewail City of Science and Technology, and KAUST.
This publication acknowledges KAUST support, but has no KAUST affiliated authors.
PY - 2019/1/24
Y1 - 2019/1/24
N2 - Optimization of the LTE network is crucial to obtain the best performance. The handover margin (HOM) and time to trigger (TTT) should be chosen so that the system will have minimum number of handovers per user per second, minimum system delay, and maximum throughput. In this paper a new handover optimization algorithm for long term evolution (LTE) network based on Q-learning optimization is presented. The proposed algorithm operates by testing different values of HOM and TTT then observes the output performance corresponding to the values of these parameters, and it eventually selects the values that produce the best performance. The proposed handover optimization technique is evaluated and compared to previous work. Q-learning achieves minimum average number of handover per user and also has maximum throughput than the fuzzy logic optimization technique.
AB - Optimization of the LTE network is crucial to obtain the best performance. The handover margin (HOM) and time to trigger (TTT) should be chosen so that the system will have minimum number of handovers per user per second, minimum system delay, and maximum throughput. In this paper a new handover optimization algorithm for long term evolution (LTE) network based on Q-learning optimization is presented. The proposed algorithm operates by testing different values of HOM and TTT then observes the output performance corresponding to the values of these parameters, and it eventually selects the values that produce the best performance. The proposed handover optimization technique is evaluated and compared to previous work. Q-learning achieves minimum average number of handover per user and also has maximum throughput than the fuzzy logic optimization technique.
UR - http://hdl.handle.net/10754/679458
UR - https://ieeexplore.ieee.org/document/8623826/
UR - http://www.scopus.com/inward/record.url?scp=85062215117&partnerID=8YFLogxK
U2 - 10.1109/MWSCAS.2018.8623826
DO - 10.1109/MWSCAS.2018.8623826
M3 - Conference contribution
SN - 9781538673928
SP - 194
EP - 197
BT - 2018 IEEE 61st International Midwest Symposium on Circuits and Systems (MWSCAS)
PB - IEEE
ER -