TY - JOUR
T1 - QoS-aware dynamic bandwidth reallocation with deadline assurance for multipath data offloading
AU - Lin, Chuan Sheng
AU - Sou, Sok Ian
N1 - Funding Information:
The work of S.-I. Sou was sponsored in part by Ministry of Science and Technology (MOST), Taiwan, under the contract number MOST 105-2628-E-006-007-MY3.
Funding Information:
The work of S.-I. Sou was sponsored in part by Ministry of Science and Technology (MOST), Taiwan, under the contract number MOST 105-2628-E-006-007-MY3 .
Publisher Copyright:
© 2019
PY - 2019/4/22
Y1 - 2019/4/22
N2 - Multipath mobile data offloading allows a user equipment (UE) to efficiently aggregate bandwidth from the cellular network and WiFi network. In deadline assurance services, users with a WiFi connection usually have more bandwidth than they require. However, those without a WiFi connection usually lack the bandwidth they need. As a result, only some of the users have a good service experience. Accordingly, this study proposes two QoS-aware bandwidth allocation algorithms for cellular network operators in which the excess cellular bandwidth allocated to QoS-satisfied sessions is reallocated to QoS-unsatisfied sessions in accordance with an Earliest Deadline First (QEDF) allocation policy or a Smallest Download Size First (QSSF) allocation policy, respectively. The effectiveness of the proposed algorithms is demonstrated by means of numerical simulations. It is shown that both algorithms successfully improve the service quality by reducing both the deadline miss ratio of the deadline assurance service and the amount of saved LTE bandwidth.
AB - Multipath mobile data offloading allows a user equipment (UE) to efficiently aggregate bandwidth from the cellular network and WiFi network. In deadline assurance services, users with a WiFi connection usually have more bandwidth than they require. However, those without a WiFi connection usually lack the bandwidth they need. As a result, only some of the users have a good service experience. Accordingly, this study proposes two QoS-aware bandwidth allocation algorithms for cellular network operators in which the excess cellular bandwidth allocated to QoS-satisfied sessions is reallocated to QoS-unsatisfied sessions in accordance with an Earliest Deadline First (QEDF) allocation policy or a Smallest Download Size First (QSSF) allocation policy, respectively. The effectiveness of the proposed algorithms is demonstrated by means of numerical simulations. It is shown that both algorithms successfully improve the service quality by reducing both the deadline miss ratio of the deadline assurance service and the amount of saved LTE bandwidth.
UR - http://www.scopus.com/inward/record.url?scp=85062471489&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85062471489&partnerID=8YFLogxK
U2 - 10.1016/j.comnet.2019.02.014
DO - 10.1016/j.comnet.2019.02.014
M3 - Article
AN - SCOPUS:85062471489
SN - 1389-1286
VL - 153
SP - 103
EP - 112
JO - Computer Networks
JF - Computer Networks
ER -