Abstract
In this paper we present an analytical framework to improve the energy consumption of mobile nodes through traffic offloading via Wireless Local Area Networks (WLANs), taking into account the energy consumption for both data transmission and network discovery operations. More specifically, we formulate an optimization problem, according to which the network scanning period is optimized in order to minimize the total energy consumption and the energy consumption per transmitted bit in a scenario where a user moves with a constant, either pedestrian or vehicular, speed along a road covered by a long range cellular network and a number of randomly deployed WLANs. The performance of the system that employs the proposed framework, which uses information on the user speed as well as on the availability and the load level of neighboring networks and performs periodic network scanning with the optimal period, is compared against a sub-optimal system that does not take into consideration the user and network context information when determining the network scanning period. According to performance evaluation results, the use of the optimal network scanning period achieves significant improvement in terms of total energy consumption, energy efficiency and network detection delay.