Mi Jeong Yang

Solving the data overload: Device-to-device bearer control architecture for cellular data offloading

The explosive growth of the mobile user population and multimedia services are causing a severe traffic overload problem in the cellular network. The Third-Generation Partnership Project (3GPP) has defined data offloading as a critical area to cope with this problem. Local IP access (LIPA) and selected IP traffic offload (SIPTO) are considered to be the typical technologies to implement cellular data offloading. Currently, device-to-device (D2D) communication has been proposed as a new dataoffloading solution. The D2D communication underlying a cellular infrastructure reduces the radio access network load as well as the core network load. The cellular network needs to establish a new type of radio bearer for D2D communications, and this bearer can provide the direct traffic path between devices. In this article, we propose a D2D bearer control architecture for D2D communications supported by a longterm evolution-advanced (LTE-A) infrastructure. This could be a feasible and efficient solution for offloading data from the cellular network.

Handover prediction strategy for 3G-WLAN overlay networks

In 3G-WLAN overlay network environments, a UE is apt to lose the WLAN signal without any notice by moving while it connects to WLAN because it can be apart from the WLAN coverage suddenly. In this case the on-going services are disrupted by unexpected link corruption. Hence, it needs to execute quickly the handover to 3G network as soon as it detects the lost of WLAN signal. However, as the 3G access has many network signaling operations additionally including IP mobility operations for service continuity, it requests more enough processing time for the operations, which increases the handover latency causing the service interruption. Therefore it is important to predict the time when the UE will get out of the WLAN coverage. By the handover prediction, the UE can prepare the connection to 3G network before breaking off the WLAN connection. This paper shows several handover prediction algorithms to reduce the handover latency by fast handover triggering. The correctness of the algorithms is compared by simulation.