The present invention relates to a context-aware video caching method and
system, and belongs to the field of mobile edge
network computing technology. First, all videos in a target
community are divided into video segments of equal length, and each
user device is divided into multiple virtual devices. Then, based on a multi-armed
slot machine algorithm, the actual capacity of all video segments in a predicted time period is effectively predicted. Finally, the video segment is used as the device of a UCKM framework, the actual capacity of the video segment is used as the capacity of the device, and the
virtual device is used as the position of the UCKM framework. With the goal of minimizing
transmission delay, the video caching strategy for the predicted time period is determined, and a video caching decision can be made in real time to minimize the average
delay of user devices that perform video caching in the
community, so as to solve the problem that the probability of different users requesting videos of different contents is uncertain, and blind caching will cause unnecessary
energy consumption and
delay, thereby improving the user
service quality.