Optimizing Handover Parameters by Q-Learning for Heterogeneous Radio-Optical Networks

Existing literature studying the access point (AP)-user association problem of heterogeneous radio-optical networks either investigates quasi-static network selection or only considers vertical handover (VHO) dwell time from optical to radio. The quasi-static assumption can result in outdated decisi...

Full description

Saved in:
Bibliographic Details
Main Authors: Sihua Shao, Guanxiong Liu, Abdallah Khreishah, Moussa Ayyash, Hany Elgala, Thomas D. C. Little, Michael Rahaim
Format: Article
Language:English
Published: IEEE 2020-01-01
Series:IEEE Photonics Journal
Subjects:
Online Access:https://ieeexplore.ieee.org/document/8903258/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849416245384839168
author Sihua Shao
Guanxiong Liu
Abdallah Khreishah
Moussa Ayyash
Hany Elgala
Thomas D. C. Little
Michael Rahaim
author_facet Sihua Shao
Guanxiong Liu
Abdallah Khreishah
Moussa Ayyash
Hany Elgala
Thomas D. C. Little
Michael Rahaim
author_sort Sihua Shao
collection DOAJ
description Existing literature studying the access point (AP)-user association problem of heterogeneous radio-optical networks either investigates quasi-static network selection or only considers vertical handover (VHO) dwell time from optical to radio. The quasi-static assumption can result in outdated decisions for highly mobile scenarios. Solely focusing on the optical to radio handover ignores the importance of dwell time for VHO from radio to optical. In this paper, we propose a flexible and holistic framework, that runs a self-optimizing algorithm at the centralized coordinator (CC). This CC resides in the LTE eNodeB and controls the handover parameters of all the visible light communication (VLC) APs under the coverage of the LTE eNodeB. Based on Q-learning approach, the algorithm optimizes the time-to-trigger (<inline-formula><tex-math notation="LaTeX">$TTT$</tex-math></inline-formula>) values for VHO between LTE and VLC. Case studies are performed to validate the considerable gain in terms of average throughput by optimizing <inline-formula><tex-math notation="LaTeX">$TTT$</tex-math></inline-formula>s. We evaluate the impact of learning parameters on the optimal throughput and convergence speed through trace-driven simulations. The simulation results reveal that the Q-learning based algorithm improves the average throughput of mobile device by 25&#x0025; when compared to the fixed <inline-formula><tex-math notation="LaTeX">$TTT$</tex-math></inline-formula> scheme. Furthermore, this algorithm is capable of self-optimizing handover parameters in an online manner.
format Article
id doaj-art-7ddc8e8a9ea240ed9d9dfbd55e9e7091
institution Kabale University
issn 1943-0655
language English
publishDate 2020-01-01
publisher IEEE
record_format Article
series IEEE Photonics Journal
spelling doaj-art-7ddc8e8a9ea240ed9d9dfbd55e9e70912025-08-20T03:33:14ZengIEEEIEEE Photonics Journal1943-06552020-01-0112111510.1109/JPHOT.2019.29538638903258Optimizing Handover Parameters by Q-Learning for Heterogeneous Radio-Optical NetworksSihua Shao0https://orcid.org/0000-0002-2831-9860Guanxiong Liu1Abdallah Khreishah2https://orcid.org/0000-0003-1583-713XMoussa Ayyash3https://orcid.org/0000-0003-0868-143XHany Elgala4https://orcid.org/0000-0002-7098-9278Thomas D. C. Little5https://orcid.org/0000-0001-5208-5358Michael Rahaim6Department of Electrical Engineering, New Mexico Tech, Socorro, NM, USADepartment of Electrical and Computer Engineering, New Jersey Institute of Technology, Newark, NJ, USADepartment of Electrical and Computer Engineering, New Jersey Institute of Technology, Newark, NJ, USADepartment of Computer and Mathematical Sciences, Lewis University, Romeoville, IL, USADepartment of Electrical and Computer Engineering, SUNY at Albany, Albany, NY, USADepartment of Electrical and Computer Engineering, Boston University, MA, USADepartment of Engineering, University of Massachusetts Boston, MA, USAExisting literature studying the access point (AP)-user association problem of heterogeneous radio-optical networks either investigates quasi-static network selection or only considers vertical handover (VHO) dwell time from optical to radio. The quasi-static assumption can result in outdated decisions for highly mobile scenarios. Solely focusing on the optical to radio handover ignores the importance of dwell time for VHO from radio to optical. In this paper, we propose a flexible and holistic framework, that runs a self-optimizing algorithm at the centralized coordinator (CC). This CC resides in the LTE eNodeB and controls the handover parameters of all the visible light communication (VLC) APs under the coverage of the LTE eNodeB. Based on Q-learning approach, the algorithm optimizes the time-to-trigger (<inline-formula><tex-math notation="LaTeX">$TTT$</tex-math></inline-formula>) values for VHO between LTE and VLC. Case studies are performed to validate the considerable gain in terms of average throughput by optimizing <inline-formula><tex-math notation="LaTeX">$TTT$</tex-math></inline-formula>s. We evaluate the impact of learning parameters on the optimal throughput and convergence speed through trace-driven simulations. The simulation results reveal that the Q-learning based algorithm improves the average throughput of mobile device by 25&#x0025; when compared to the fixed <inline-formula><tex-math notation="LaTeX">$TTT$</tex-math></inline-formula> scheme. Furthermore, this algorithm is capable of self-optimizing handover parameters in an online manner.https://ieeexplore.ieee.org/document/8903258/HandoverQ-learningheterogeneous networkvisible light communicationparameter optimization.
spellingShingle Sihua Shao
Guanxiong Liu
Abdallah Khreishah
Moussa Ayyash
Hany Elgala
Thomas D. C. Little
Michael Rahaim
Optimizing Handover Parameters by Q-Learning for Heterogeneous Radio-Optical Networks
IEEE Photonics Journal
Handover
Q-learning
heterogeneous network
visible light communication
parameter optimization.
title Optimizing Handover Parameters by Q-Learning for Heterogeneous Radio-Optical Networks
title_full Optimizing Handover Parameters by Q-Learning for Heterogeneous Radio-Optical Networks
title_fullStr Optimizing Handover Parameters by Q-Learning for Heterogeneous Radio-Optical Networks
title_full_unstemmed Optimizing Handover Parameters by Q-Learning for Heterogeneous Radio-Optical Networks
title_short Optimizing Handover Parameters by Q-Learning for Heterogeneous Radio-Optical Networks
title_sort optimizing handover parameters by q learning for heterogeneous radio optical networks
topic Handover
Q-learning
heterogeneous network
visible light communication
parameter optimization.
url https://ieeexplore.ieee.org/document/8903258/
work_keys_str_mv AT sihuashao optimizinghandoverparametersbyqlearningforheterogeneousradioopticalnetworks
AT guanxiongliu optimizinghandoverparametersbyqlearningforheterogeneousradioopticalnetworks
AT abdallahkhreishah optimizinghandoverparametersbyqlearningforheterogeneousradioopticalnetworks
AT moussaayyash optimizinghandoverparametersbyqlearningforheterogeneousradioopticalnetworks
AT hanyelgala optimizinghandoverparametersbyqlearningforheterogeneousradioopticalnetworks
AT thomasdclittle optimizinghandoverparametersbyqlearningforheterogeneousradioopticalnetworks
AT michaelrahaim optimizinghandoverparametersbyqlearningforheterogeneousradioopticalnetworks