Key points are not available for this paper at this time.
The study item: "Feasibility Study on LTE-based V2X Services", approved at 3GPP TSG RAN #68, has aroused the interest in the study of LTE assisted vehicle-to-vehicle (V2V) and vehicle-to- infrastructure (V2I) communications in the networks of connected vehicles. By deploying the direct device-to-device (D2D) technology of traditional cellular networks into the V2X (including both V2V and V2I) communications, performance improvements can be expected, such as better reliability, lower latency, and more efficient content sharing. This paper investigates the content sharing problem in the D2D based V2X communication networks. With both vehicles and eNBs carrying multiple different classes of data, this work studies how to optimize the information exchanged within the network, and in the mean time to guarantee the system quality of service (QoS) requirements. By jointly considering the data diversity and link quality, the interactions between vehicles/eNBs, or in other words, the V2V and V2I link scheduling, is modeled as the stable fixture (SF) matching game. Different from traditional D2D communications, where each node is limited to one link, we allow multiple V2X connections for each vehicle to further optimize the content sharing. More specifically, the formations of such V2X links are independent from each other, thus more flexible than the conventional clustering formation in the Vehicular ad hoc networks (VANETs). The SF game is solved by the proposed Irving's stable fixture (ISF) algorithm. Its advantages over some heuristics are demonstrated through simulation experiments.
Gu et al. (Thu,) studied this question.