Energy Efficient Prediction Based D2D-Assisted Pushing in Heterogeneous Ultra-dense Networks

Abstract

Both the low latency and high energy efficiency are expected in the 5th generation mobile communication system (5G). The content prediction is believed as a promising way to reduce the latency. However, pushing a large amount of content will result in low energy efficiency. In order to get higher energy efficiency, we propose the prediction based D2D-assisted pushing in Heterogenous Ultra-dense Networks (H-UDNs). The D2D relay-user equipment can assist the pushing process of user equipments who are occupied with other services at that time, which can benefit on energy efficiency and latency reduction. Moreover, a new metric of energy efficiency for prediction is proposed in this paper called the Ratio of Energy to Prediction Success Rate (REPSR). Specifically, it can describe the energy efficient characteristics of prediction. Therefore, we can improve the energy efficiency and reduce the latency through the energy efficient prediction based D2D-assisted pushing.