A Machine-Learning-Based Handover Prediction for Anticipatory Techniques in Wi-Fi Networks

Handover and blind spots in Wi-Fi networks generate temporary interruptions of connection between the devices and the access point, with major quality degradation, for example to video streaming. In this paper we propose a technique to predict the event of handover and blind spots in order to allow the implementation of anticipatory techniques, where connection resources are reallocated or video buffers are filled with low-definition video frames before the connection gets lost. The prediction is based on a machine-learning approach, where the received signal strength indicator (RSSI) is monitored and an upcoming handover is recognized by the pattern of the RSSI over time. Since a number of impairments (different paths followed by the user, different movement speed, fading, noise) affect the RSSI evolution, we resort to a neural-network to learn the peculiarities of each handover and solve the pattern recognition problem.