The preemptive caching policy learning system considering the video quality of the mmWave vehicle network based on deep reinforcement learning according to the embodiment includes an information storage unit receiving and storing vehicle information and base station information, and performing deep reinforcement learning using the provided information. And a control unit for allocating the quality information of the video data and the capacity of the video data to the base station to be connected to the vehicle based on the in-depth reinforcement learning unit and the learned information. In the embodiment, by learning using the DDPG learning algorithm, there is an effect that big data can be seamlessly transmitted in a large-scale vehicle network.
展开▼