Abstract
In this paper, we consider a hybrid user association (HUA) problem for load balancing of multitier cellular networks. The proposed hierarchical HUA approach builds on a combination of decentralized user association (DUA) and auxiliary intervention of a central control unit (CCU). A major challenge with the CCU intervention is the time interval determined by a selected CCU control cycle during which the DUA must accept all users that satisfy the prevailing association criterion while proactively mitigating potential resource depletions. Consequently, the primary focus of this work is on relating the control cycle of the CCU intervention with the incipient resource depletions, according to a maximum allowed resource depletion probability. By uniquely combining a set of mathematical tools from stochastic geometry and queueing theory, we present a novel HUA method which evaluates the association bias values of the DUA according to a CCU-optimized load vector and enables tier-based resource depletion probability provisioning over finite control cycles. The trade-offs between the proposed HUA method and the standard DUA approach are demonstrated via network simulations with flow-level spatiotemporal dynamics.
| Original language | English |
|---|---|
| Title of host publication | 2019 IEEE 89th Vehicular Technology Conference, VTC Spring 2019 |
| Publisher | IEEE Institute of Electrical and Electronic Engineers |
| Number of pages | 6 |
| ISBN (Electronic) | 978-1-7281-1217-6 |
| ISBN (Print) | 978-1-7281-1218-3 |
| DOIs | |
| Publication status | Published - 2019 |
| MoE publication type | A4 Article in a conference publication |
| Event | 89th IEEE Vehicular Technology Conference, VTC Spring 2019 - Kuala Lumpur, Malaysia Duration: 28 Apr 2019 → 1 May 2019 |
Publication series
| Series | IEEE Vehicular Technology Conference Papers |
|---|---|
| Volume | 2019-April |
| ISSN | 1090-3038 |
Conference
| Conference | 89th IEEE Vehicular Technology Conference, VTC Spring 2019 |
|---|---|
| Country | Malaysia |
| City | Kuala Lumpur |
| Period | 28/04/19 → 1/05/19 |
Fingerprint
Keywords
- Load balancing
- Network heterogeneity
- Predictive congestion control
- Queueing theory
- Stochastic geometry
Cite this
}
Hybrid user association with proactive auxiliary intervention for multitier cellular networks. / Anttonen, Antti; Mammela, Aarne; Chen, Tao.
2019 IEEE 89th Vehicular Technology Conference, VTC Spring 2019. IEEE Institute of Electrical and Electronic Engineers , 2019. (IEEE Vehicular Technology Conference Papers, Vol. 2019-April).Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review
TY - GEN
T1 - Hybrid user association with proactive auxiliary intervention for multitier cellular networks
AU - Anttonen, Antti
AU - Mammela, Aarne
AU - Chen, Tao
PY - 2019
Y1 - 2019
N2 - In this paper, we consider a hybrid user association (HUA) problem for load balancing of multitier cellular networks. The proposed hierarchical HUA approach builds on a combination of decentralized user association (DUA) and auxiliary intervention of a central control unit (CCU). A major challenge with the CCU intervention is the time interval determined by a selected CCU control cycle during which the DUA must accept all users that satisfy the prevailing association criterion while proactively mitigating potential resource depletions. Consequently, the primary focus of this work is on relating the control cycle of the CCU intervention with the incipient resource depletions, according to a maximum allowed resource depletion probability. By uniquely combining a set of mathematical tools from stochastic geometry and queueing theory, we present a novel HUA method which evaluates the association bias values of the DUA according to a CCU-optimized load vector and enables tier-based resource depletion probability provisioning over finite control cycles. The trade-offs between the proposed HUA method and the standard DUA approach are demonstrated via network simulations with flow-level spatiotemporal dynamics.
AB - In this paper, we consider a hybrid user association (HUA) problem for load balancing of multitier cellular networks. The proposed hierarchical HUA approach builds on a combination of decentralized user association (DUA) and auxiliary intervention of a central control unit (CCU). A major challenge with the CCU intervention is the time interval determined by a selected CCU control cycle during which the DUA must accept all users that satisfy the prevailing association criterion while proactively mitigating potential resource depletions. Consequently, the primary focus of this work is on relating the control cycle of the CCU intervention with the incipient resource depletions, according to a maximum allowed resource depletion probability. By uniquely combining a set of mathematical tools from stochastic geometry and queueing theory, we present a novel HUA method which evaluates the association bias values of the DUA according to a CCU-optimized load vector and enables tier-based resource depletion probability provisioning over finite control cycles. The trade-offs between the proposed HUA method and the standard DUA approach are demonstrated via network simulations with flow-level spatiotemporal dynamics.
KW - Load balancing
KW - Network heterogeneity
KW - Predictive congestion control
KW - Queueing theory
KW - Stochastic geometry
UR - http://www.scopus.com/inward/record.url?scp=85068981618&partnerID=8YFLogxK
U2 - 10.1109/VTCSpring.2019.8746686
DO - 10.1109/VTCSpring.2019.8746686
M3 - Conference article in proceedings
AN - SCOPUS:85068981618
SN - 978-1-7281-1218-3
T3 - IEEE Vehicular Technology Conference Papers
BT - 2019 IEEE 89th Vehicular Technology Conference, VTC Spring 2019
PB - IEEE Institute of Electrical and Electronic Engineers
ER -