Dynamic Connected Graph-based Content Transmission Strategy Researches In Internet Of Vehicles

With the development of Internet of Things,the number of mobile devices is growing at an exponential rate.Currently,due to the dynamic topologies and the unstable content transmission links in Internet of Vehicles(IoVs),low latency and high delivery demands are facing enormous challenges.This work focuses on the content transmission in IoVs based on dynamic connected graphs.The main contributes are summarized as follows:(1)In order to extract and apply the social relationships of mobile nodes in content transmission in IoVs,this work extracts the social relationships of IoVs by triangle primitive structure and proposes a triangle motif based clustering algorithm in social IoVs to prove the availability and efficiency of triangle motif in extracting social relationships,which can be used to improve the efficiency of content transmission in IoVs.(2)For the Device-to-Device(D2D)content transmission in the IoVs,this paper proposes a deep learning-based data transmission strategy by exploring tri-relationship among vehicles.The method extracts the various features from heterogeneous IoVs,with which a weighted device network can be obtained.Then,we consider the social and physical layers of the constructed network.Hence,the efficiency of D2 D communication can be improved.(3)For two transmission links,i.e.,Vehicle-to-Vehicle(V2V)and Vehicle-to-Road Side Unit(V2R),we proposes a fog-based content transmission scheme with collaborative filtering.Based on the real-time environmental perception of fog computing,the proposed strategy perceives and measures the similarities of trajectories and real-time encounter situations of mobile terminal nodes and establishes a stable V2 V link.For V2 R,this work models vehicle file downloading procedures by a two-dimensional Markov model.Furthermore,the popularity and availability file vectors are also defined based on file requests in IoVs.Then,files can be update in RSUs.Then the efficiency of V2 R can be improved.Simulation experiments are carried on the real vehicular dataset.The experimental results demonstrate that the proposed method can improve the efficiency of content transmission in IoVs.