Data Traffic Collection And Modeling For ICN-based Vehicular Networks

The vehicular networks and the Information Centric Networks(ICN)have the natural connection and similarity.The introduction of ICN technique in vehicular networks can improve the efficiency of information transmission.However,the introduction of ICN technique into the vehicular networks will cause changes in the data forwarding protocols and the related mechanisms such as routing,packet forwarding,data caching,application types,data acquisition methods,etc.This changes greatly the characteristics of the data traffic of vehicular networks.At the same time,with the development of the vehicular networks,large numbers of vehicle terminals and mobile devices with various types of sensors are connected to the network,which generates huge amount of data traffic.This will bring unpredictable impact on the current network infrastructure.Therefore,it is necessary to model the data traffic in ICN-based vehicular networks.This will benefit the resource allocation of the network infrastructure and the guarantee of applications’ performance.The caching scheme is helpful for the study of the impact on different caching strategies on the traffic characteristics.In particular,the caching scheme considers the popularity of information and the vehicle density.These two factors are also two important factors affecting the traffic characteristics.On the basis of analyzing the in-network caching technologies and characteristics of data traffic in ICN-based vehicular networks,we designs and implementations an implicit cooperative caching scheme based on information popularity.The simulation results show that the scheme reduces the times of information caching and cache redundancy and saves the whole network storage space.The scheme has little effect on the average response time of obtaining information from the network.According to the above,a data traffic generate system has been designed and implemented in the simulation environment based on the vehicles’ trace data collected from the real world,which also considers the types of applications,the in-network caching schemes.Then,we perform network simulation,generate and collect the traffic in the ICN-based vehicular network.In order to model the data traffic of the ICN-based vehicular networks,we propose a nested Hidden Markov Model based on wavelet modulation.The model decomposes the complex ICN-based vehicular network traffic into different oscillation states with wavelet transform,and then models the different oscillation states using hidden Markov model,forming a nested hidden Markov model.The proposed model has been evaluated in terms of accuracy,adaptability and complexity.The results show that the model can well describe the characteristics of the data traffic in ICN-based vehicular networks.