Energy-Efficient Traffic Offloading For Mobile Users In Heterogeneous Wireless Networks

Due to the popularity of mobile Internet access, wireless networks are facing the challenge of explosive data traffic generated by various services and applications. However, it has been recognized that mobile user equipments(UEs) cost a lot of energy consumption because of the randomness of mobile network and the congestion caused by the randomness of UEs. Many studies about WiFi offloading have been proposed to alleviate the traffic of cellular network. A delayed offloading framework have been proposed where the traces indicate that WiFi networks can offloaded about 50% of the total mobile data traffic. Although much research has studied traffic offloading strategies, little has considered user mobility and energy efficiency in large-scale heterogeneous wireless networks(HWNs). In this paper, we consider a large-scale two-tier heterogeneous wireless network modeled using stochastic approaches. We study to minimize total energy consumption of mobile user equipment by designing an algorithm that determines how UEs switch between different networks along their traveling trajectory.In this paper, we will study a problem of minimizing total energy consumption of UEs while satisfying their data transmission demands. To solve this problem, we need to design a data offloading algorithm to decide when UEs should switch between WiFi and cellular networks. Firstly, we model a data offloading problem for mobile users in a large-scale heterogeneous network consisting of both WiFi and cellular networks. We prove that the expected time of WiFi coverage of UEs is irrelevant to their trajectories. Secondly, we proposed the energy minimization model. Based on the above important prove, we formulate the energy minimization problem as a quadratic programming problem that can be quickly solved by existing mathematical tools.Finally, we conduct extensive simulation to evaluate the performance of our proposed algorithm. The results show that it can save up to 34% energy in dense networks while still maintains an acceptable completion rate well above 94%.