A Trajectory Inference-based Technique for Energy Efficient Store-and-Forward Technology

The wireless network infrastructure is critical for many applications, often operating over energy-sensitive networks with large communication delays. The store-and-forward technology used in Delay/Disruption Tolerant Networking (DTN) can minimize this problem. However, DTN requires efficient energy consumption technology for increasing the mobile node lifetime, especially when the connection opportunities among nodes change over time. This paper proposes an energy-saving technique on store-and-forward technology that saves energy by employing a node trajectory inference model based on machine learning for communication control. Experimental results indicate more than 47% of energy-saving on data communication applying the proposed trajectory inference model.