Search

With the proliferation of mobile devices in recent years, the volume of global mobile data traffic has grown exponentially. As a result, effective approaches for mobile data offloading are urgently required. In the past few years, two main Wi-Fi offloading techniques, namely, opportunistic offloading and delayed offloading, have been widely discussed. A promising L4-approach known as multipath transmission control protocol has recently been standardized, which allows a single transport connection to use multiple paths simultaneously over multiple interfaces, which then calls for a new Wi-Fi offloading model that considers discontinuous Wi-Fi connection periods. This paper presents an analytical model for multipath Wi-Fi offloading in deriving the aggregate offloading time via an alternative path for the use of multipath offloading. Then, the respective offloading volume and ratio are computed. Our theoretical results are supported by extensive simulations. We show how this model further assists the cellular operators to make better decision in resource allocation. Compared with opportunistic offloading, multipath offloading results in a significantly improved performance in the support for a guaranteed minimum bit rate. [ABSTRACT FROM AUTHOR]

The performance of single-segment short messaging service (SMS) (i.e., the message sizes are smaller than 140 octets) has been intensively investigated. On the other hand, multisegment messages (with sizes larger than 140 octets) are seldom evaluated in the literature. This paper proposes analytical models to study two transmission policies for multisegment short messages. We show how to improve the performance of multisegment short message delivery by selecting appropriate transmission parameters: the retransmission interval and the maximum number of transmissions. The proposed models are validated by measured data collected from the largest telecom operator in Taiwan over a period of six months. We provide useful guidelines to configure parameters for SMS transmission policy. Numerical examples showed that the performance can be improved by over 20% in terms of successful delivery ratio. [ABSTRACT FROM AUTHOR]