Research On Online Adaptive Routing Algorithms In Software-Defined Vehicular Networks

Vehicular networks are crucial for the construction of intelligent transportation systems,which ensure the information transmission to provide some basic services.In the past two decades,vehicular ad hoc network(VANET)has been the focus of researchers.This kind of self-organized networks can provide low-latency communications,but the lack of centralized control and global information may lead to low efficiency of network communication.Software-defined vehicular network(SDVN)has been a promising technology for nextgeneration vehicular networking.Compared with conventional centralized SDVN,hybrid SDVN combines the centralized control of SDVN and self-organized distributed routing of VANET to mitigate the burden on the central controller caused by frequent uplink and downlink transmissions.Although a wide variety of routing protocols have been developed for vehicular networks,these protocols lack flexibility.To address this problem,this paper proposes an efficient online sequential learning-based adaptive routing scheme,namely,penicillium reproduction-based online learning adaptive routing scheme(POLAR)for hybrid SDVN.This new scheme can dynamically select a routing strategy for a specific traffic scenario by learning traffic features.Specifically,this paper firstly applies an area coding system to divide the large geographical area into multiple grids,which facilitates the collection and processing of realtime traffic data for regional management in SDVN controller.Secondly,a new penicillium reproduction algorithm(PRA)with outstanding optimization capabilities is designed to improve the learning effectiveness of online sequential extreme learning machine(OS-ELM).Finally,POLAR is deployed in controllers to generate routing decision-making model.Based on real-time feature data,this scheme can choose the optimal routing strategy for a specific area.The results of a large number of simulations show that POLAR is superior to a single traditional routing protocol in terms of packet delivery ratio and average end-to-end delay.