Content-Prefetching and Broadcast Scheduling in Vehicular Networks with a Realistic Channel Model

Prefetching content at the road side units (RSUs) helps improve the throughput of vehicular networks by reducing the average data access latency from the remote servers and reaping the benefits of popular content reuse to serve multiple interested users from the caches. In the literature, it is commonly assumed that a single constant physical layer data rate is used for communication between RSUs and vehicles, as a result, there is a lack of understanding of the impact of data rate selection on the system performance. In this paper, we study the joint content prefetching and broadcast scheduling in vehicular networks with the support of RSUs. We adopt different data rates for broadcasting different contents. We consider a more realistic channel model where the successful data reception probability monotonically decreases with the increasing distance of the vehicles from the RSU. This is in stark contrast to traditional (unrealistic) models that assume as a single threshold range for data reception. We formulate the problem as an integer linear programming problem and propose heuristic prefetching and scheduling algorithms to solve the problem efficiently. Through simulation, we show that: i) broadcasting with adaptive data rates is better than the using of a single uniform rate; ii) using the predicted vehicular trajectories and the content popularity together yields better results than simply relying on the content popularity for content prefetching; and finally, iii) using the continuous data reception channel model proves to achieve better performance than just relying on a single reception threshold, especially when the dwelling time of vehicles in the vicinity of the RSU is short, compared to the period to broadcast all prefetched data on the RSUs.