Anticipatory Mobility Management by Big Data Analytics for Ultra-Low Latency Mobile Networking

Massive deployment of autonomous vehicles, unmanned aerial vehicles, and robots, brings in a new technology challenge to establish ultra-low end-to-end latency mobile networking to enable holistic computing mechanisms. With the aid of open-loop wireless communication and proactive network association in vehicle-centric heterogeneous network architecture, anticipatory mobility management relying on inference and learning from big vehicular data plays a key role to facilitate such a new technological paradigm. Anticipatory mobility management aims to predict APs to be connected in the next time instant and in a real-time manner, such that ultra-low latency downlink open-loop communication can be realized with proactive network association. In this paper, we successfully respond this technology challenge using big data analytics with location-based learning and inference techniques, to achieve satisfactory performance of predicting APs. Real vehicular movement data have been used to verify that the proposed prediction methods are effective for the purpose of anticipatory mobility management and thus ultra-low latency mobile networking.