Q-Learning-based Prediction of Channel Quality after Handover in Mobile Networks

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F14%3A00223318" target="_blank" >RIV/68407700:21230/14:00223318 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Q-Learning-based Prediction of Channel Quality after Handover in Mobile Networks

Popis výsledku v původním jazyce

To avoid call drops after handover due to unavailability of radio resources at a target handover cell, call admission control procedure reserves a specific amount of resources for users performing handover to this cell. If a high amount of resources is reserved, the available capacity for users served by the cell is lowered. Contrary, if a low amount of resources is booked for users entering the new cell, handover cannot be performed and user's connection is dropped. To optimize the amount of reserved resources, we propose an algorithm for prediction of channel quality between the user and the target cell after completing handover to the target cell. The algorithm is based on the knowledge of handover hysteresis and on decomposition of overall interference caused by other cells in the network. The prediction accuracy is tuned by correction parameter, which is dynamically set based on Q-learning approach. As the results show, the proposed algorithm with learning improves the efficiency

Název v anglickém jazyce

Q-Learning-based Prediction of Channel Quality after Handover in Mobile Networks

Popis výsledku anglicky

To avoid call drops after handover due to unavailability of radio resources at a target handover cell, call admission control procedure reserves a specific amount of resources for users performing handover to this cell. If a high amount of resources is reserved, the available capacity for users served by the cell is lowered. Contrary, if a low amount of resources is booked for users entering the new cell, handover cannot be performed and user's connection is dropped. To optimize the amount of reserved resources, we propose an algorithm for prediction of channel quality between the user and the target cell after completing handover to the target cell. The algorithm is based on the knowledge of handover hysteresis and on decomposition of overall interference caused by other cells in the network. The prediction accuracy is tuned by correction parameter, which is dynamically set based on Q-learning approach. As the results show, the proposed algorithm with learning improves the efficiency

Druh

D - Stať ve sborníku

CEP obor

JA - Elektronika a optoelektronika, elektrotechnika

OECD FORD obor

—

Projekt

<a href="/cs/project/GPP102%2F12%2FP613" target="_blank" >GPP102/12/P613: Predikční algoritmy pro efektivní řízení mobility uživatelů v bezdrátových sítích</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2014

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

IEEE 25th Annual International Symposium on Personal, Indoor and Mobile Radio Communications

ISBN

978-1-4799-4912-0

ISSN

2166-9589

e-ISSN

—

Počet stran výsledku

6

Strana od-do

1359-1364

Název nakladatele

IEEE Conference Publications

Místo vydání

Piscataway

Místo konání akce

Washington D.C.

Datum konání akce

2. 9. 2014

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—