Cooperative Caching in a Content-Centric Network for High-Definition Map Delivery

In autonomous driving, High-Definition (HD) maps are deemed to be indispensable in making up the shortages of the sensor-based perception system of vehicles. Since the contents in HD maps are heterogeneous, complex, and latency-sensitive, the cloud-based delivery method used for lightweight and simple digital navigation maps cannot be applied in the era of HD maps. As the content-centric network (CCN) provides a latency-efficient paradigm for content distribution by decoupling content from the host and pushing contents nearby the users, we propose a CCN-based cooperative caching, computing, and request scheduling scheme to support the dissemination of HD map contents. We firstly analyze and clarify the service of HD map dissemination in terms of network requirements. Based on that, we formulate a joint edge caching, computing, and request scheduling problem that aims to minimize the total content access latency of vehicles as an Integer Linear Program (ILP). To conquer the complexity, we decompose the original problem into a cache placement problem and a request scheduling problem, which can be solved in polynomial time, respectively. Finally, we carry out numerical simulations to verify the effectiveness of our proposed scheme.