Cooperative Downloading in Vehicular Heterogeneous Networks at the Edge

Abstract

Connected and automated driving vehicles are expected to generate an increasing amount of data traffic, possibly overloading vehicle-to- network (V2N) communication infrastructure in the long run. In this paper, we investigate the potential of local collaboration among vehicles to mitigate the load on V2N (e.g., cellular) communication networks. Vehicles in the vicinity use vehicle-to-vehicle (V2V) communications to form a group, called vehicular micro cloud, and each of them downloads a subset of data segments that comprise an original data content. The downloaded data segments are cached and shared with other group members by way of V2V networks. This enables the group of vehicles to collectively serve as a virtual content delivery server, which complements cloud / edge computing infrastructure. In order to maximize the benefit of cooperation, we design a light-weight local coordination mechanism for vehicles to agree on non-overlapping subsets of data segments that they request from a remote server. Our simulation results show that coordination among vehicles improves the efficiency of cooperative download, reducing the data traffic on cellular networks.