Hawkes-based Collaborative Caching and Resource Allocation in Internet of Vehicles

With the exponential growth of network traffic, intelligent vehicles that can access multimedia content are also faced with huge pressure of traffic. Therefore, a collaborative caching and resource allocation framework with updating content popularity based on Hawkes processes of Internet of vehicles was proposed. A collaborative caching and resource allocation strategy in roadside units and intelligent vehicles was studied. Considering that the update period of content caching was much larger than the change period of channel conditions, a double timescale model was proposed. First, while considering the freshness and timeliness of the content request, the popularity based on the historical content request record was updated by using the method of Hawkes process. Then, the data transmission throughput, caching energy consumption of roadside units and vehicles collaborative strategy was modeled with the goal to maximize the caching benefit of edge devices, and deep reinforcement learning was used to solve the optimization problem. Simulation results show that the proposed strategy can get higher benefit than other strategies.