A density-aware probabilistic interest forwarding method for content-centric vehicular networks

Abstract

Vehicular networks, compared to other wireless networks, face particular challenges due to the rapidly changing topology and intermittent connections. By eliminating the need to establish and maintain an end-to-end connection, Content-Centric Network (CCN) model has recently become an appropriate solution to meet the challenging demands of vehicular network communications. In this kind of network, the basic method of forwarding interest packets is flooding. This approach will result in excessive redundancy, serious contention, and collision to which it is referred as the broadcast storm problem. In this article, a probabilistic strategy is proposed to alleviate the impact of the broadcast storm on interest forwarding in content-centric vehicular networks. In this density-aware approach, each vehicle dynamically computes the probabilities based on the number of existing neighbors. A local density approximation method is presented, which uses the information provided by the newly modified Pending Interest Table (PIT) entries. Moreover, some time-based techniques are employed to give priority to potential forwarders. The simulation results indicate that the proposed work outperforms the basic CCN. With on average 40% lower network load and 65% fewer interests compared to the basic CCN, it shows only an overall 6% decrease in reachability.